Context and Results

Context: A limited archaeological excavation was carried out by GUARD on the site of major medieval drain at Paisley Abbey. This was the first excavation to investigate the deposits surrounding the drain.

Results: The remains of masonry wall and stone foundation were uncovered to the north-west of the trench that may be contemporary with the drain and could therefore be part of the monastic precinct

Notes: A limited archaeological excavation was carried out by GUARD on the site of major medieval drain at Paisley Abbey. The project was funded by Renfrewshire Council and was supported by Renfrewshire Local History Forum who provided both funds and eager volunteers. Previous work on the drain had concentrated on the archaeological deposits and artefacts recovered from inside the drain so this was the first excavation to investigate the deposits surrounding the drain. Although deep demolition layers and recent levelling deposits were identified over most of the trench there were areas of undisturbed archaeological deposits especially in the deeper sections of the trench. The remains of masonry wall and stone foundation were uncovered to the north-west of the trench that may be contemporary with the drain and could therefore be part of the monastic precinct Part of the top of the roof and north exterior wall of the drain were also exposed two distinct sections of stonework suggest that the drain may have been built in two phases or was possibly repaired certainly investigations within the drain itself had demonstrated different phases of building. In addition to the medieval structures and deposits relating to the drain there were also more recent structures and walls from the buildings and tenements that occupied the site in the 19th and early 20th century.
Draft DES Entry from GUARD DSR

The general aims of the project were two-fold:
• to assess the survival of archaeological deposits adjacent to the drain that may also provide information regarding the construction of the drain and the medieval monastic precinct
• to investigate the archaeological deposits in an area that can be used to improve access to the drain.

The aim was to obtain as much information regarding the construction of the drain and the surrounding deposits while causing the least amount of disturbance to these deposits. To ensure that this was the case excavation started in area of known disturbance to the west of the manhole. It was known that the manhole was approximately 1 m above the top of the drain. Previous work also established that the drain was blocked immediately to the west of the manhole with little indication as to how the drain continued towards the river.

The archaeological works consisted of a small open excavation initially over an area c 8 m by 6 m with the manhole in the south east corner but this was stepped in by 1 m for safety reasons and the area available for excavation reduced. Once the turf had been removed by Renfrewshire Council Parks department the topsoil and overburden was removed by machine under close archaeological supervision and then excavation continued by hand. Topsoil and spoil from the excavation was stored on top of polythene sheets in the area of the main footpath to the west of the trench within the site compound. Exposed surfaces and deposits were cleaned by trowel. Archaeological features were recorded in plan by measured drawing at a scale of 1:20 and by digital and monochrome print photography. Features were half-sectioned, with fills removed and sampled where appropriate. Sections were recorded by photograph, by measured drawing at a scale of 1:10 or 1:20 and by written description. Conditions for the work after an initial couple of very wet days were generally very good.

Results
First Step The trench was located to the west and north of the manhole currently used to access the medieval drain, the trench was 8 m north/south by 6 m east/west with the manhole in the south-west corner. Based on the results of other work in the area it was known that there had been a lot of disturbance and landscaping on the site when the 19th century tenements were demolished and the area landscaped. Therefore the decision was taken to remove the overburden from the landscaping by machine under archaeological supervision.

Work started at the southern end of the trench closest to the river, a one metre wide strip was left due to the proximity of the trees and this section was hand excavated to minimise disturbance to the root systems. The topsoil (001) was only 0.2 m thick and was removed to reveal light brown heavy clay (002) that covered most of the southern and western side of the trench and was 0.4 m thick. In the southern half of the trench the clay layer (002) overlay a deep deposit of cinders and ash with occasional large angular stones (006) again up to 0.4 m thick, these two layers may have been brought in as levelling material when the landscaping took place.

To the north the upper layers were more disturbed and consist of demolition material and the remains of service trenches. Here the clay layer covered a more mixed layer 0.35 m thick containing bricks, roofing slate and rubble as well as fragments of a ceramic sewage pipe (007). Beneath 007 was a discrete layer of brick and roofing slate up to 0.25 m thick. The remains of a masonry wall were exposed in section (015) the foundation trench for this wall had been cut from a higher level and had been sealed by clay layer 002. The remains of the masonry wall had been covered by a layer of silt and ash (034) which in turn was covered by a layer of sand (033), these layers presumably relate to the demolition of the wall. To the north of the wall below layer 008 was a thin deposit of sand (035) that sealed a service trench (010) containing a cast iron pipe that ran diagonally across the trench towards concrete base (005), the pipe had been broken before it reached the concrete so it was not possible to determine whether the pipe had originally ran into the drain/manhole.

To the south of wall (015) the ash layer (006) sealed an thin deposit of silty sand (032) this sealed an extensive deposit of black silt with few inclusions (009) up to 0.4 m thick that covered the whole of the southern half of the trench, again this may be a deliberate levelling deposits. Below this was an extensive layer that covered most of the western half of the trench and seemed to mark the end of modern disturbance. This layer (012) was 0.2 m thick and consisted of grey/brown silt with inclusions of sand and mortar with occasional fragment of tile or brick along with bottle glass and nineteenth century pottery.

In the north-east corner of the trench the upper deposits were quite different from the south-west part of the trench, a stone built wall (003) was uncovered immediately below the topsoil and appeared to continue beneath the footpath to the north, another wall (004) joined 003 at right angles and ran under the eastern baulk leaving the small corner of a building within the trench. The material contained by these two walls (016) consisted of demolition debris including bricks, slate mortar and rubble and was 0.3 m thick. Below this was a more compact grey/black silt layer (017) that contained several sherds of orange ceramic flower pots, due to constraints of time no further work was done in this area. The area to the west of wall 003 along the north trench edge was quite restricted as this side was also stepped in for safety reason but the main upper demolition deposits (007) and (010) extended over this area where it overlay a mid to dark brown silt layer (018). The foundation courses of wall 003 were cut into this layer and it may have been a garden soil. A deposit of bright orange burnt clay or possibly daub was uncovered immediately adjacent to and parallel to wall 003 and was initially interpreted as some kind of foundation course but further investigation revealed that this material was actually within 018 and not directly joined to wall 003 although it could still be part of the ground preparation for the wall or a type of damp course.

As excavation continued next to the manhole that provides access to the medieval drain it was revealed that the manhole was built on top of a large concrete base (005) that had been re-enforced with steel beams, looking down the inside of the manhole confirmed that the steel beams were sitting directly over the masonry roof of the drain and confirmed that the concrete had been applied directly over the medieval masonry. The concrete was exposed on two sides of the brick manhole and extended out from the manhole for approximately 1 m. A section of concrete forming the north-west corner of the base was broken and it was possible to remove this by machine to allow further investigation of the deposits below.

Second Step - As the level of layers 012 and 018 which covered most of the trench and the concrete base of the manhole were at a depth of approximately 1.2 m for safety reasons a one metre wide step was left along the south, west and north trench edges while along the east trench edge walls 003 and 004 meant that little further excavation could take place and they formed the safety step. Therefore further excavation was restricted to an area in the centre of the trench 3.2 m north/south by 2 m east/west. Initially a baulk was left in the middle of this area to maintain stratigraphic control and to allow excavation to continue to the south while features 011 and 013 were investigated.

This central area was cleaned and excavated by hand, immediately to the north of the manhole the concrete base overlay a large area of flat mortared stones (020) and a very loose deposit of white sand (011) that contained a few fragments of stones further cleaning to the west revealed a mixed deposit containing red and orange sand silt (013) that had possibly been burnt, this layer overlay (018) and the white sand (011). The extent of the burnt material was uncovered to reveal a sub-oval feature 1.10 m by 0.5 m this was half sectioned. Excavation revealed that the deposit was very mixed and a maximum of 0.18 m thick. Due to the mixed nature of the deposit the edges were not always very clear and patches and thin lenses of burnt material were uncovered in the deposits below. Although a fragment of roof tile with a knib for hanging the tile was recovered from (013) along with a small sherd of medieval pottery these deposits probably represent demolition and levelling.

Once 013 was excavated the remainder of layer 012 and 018 was removed to the south and east to reveal the full extent of the white sand (011) this seemed to curve in an arc and did not extend to the south of the central baulk (located to the north of section line B-B’ on plan) where excavation had not encountered the sand apart from a small area next to the baulk. Further excavation revealed that several stones exposed below (013) were in fact part of a much larger block of mortared masonry (025) that marked the extent of the white sand to the west. As the sand was removed it revealed the full extent of the masonry. The sand was filling a cut (041) that was visible mainly at the eastern extent where it had been cut through silt deposits (023) against the outside wall face of the medieval drain and into the underlying deposits (024) below the sand. Deposit 024 was largely confined to the area between 025 and the drain (028/030) and consisted of grey/brown silt with sand and mortar and was similar to deposits 021/023. Animal bone, shell, tile and one sherd of medieval pottery was recovered from the upper material of this deposit as due to time constraints this was not fully excavated.

To the west of the sand was the exposed masonry (025) this consisted of a large sub-triangular fragment of mortared masonry 1.35 m east/west by 1.7 m north/south and was up to 0.35 m thick. The masonry had been built on a large flat masonry foundation slab (029) that extended 0.8 m to the west of the masonry and ran into the baulk. This foundation to the west of the masonry fragment was covered by a thin layer of sand and mortar (031) which in turn was covered by a deep deposit of mid brown silt (038) with mortar similar to 014 but with patches of orange burnt material that may be part of (013), this deposit was sealed by the main layer 012. The masonry (025) comprised a loose light brown mortar with large angular stones up to 0.4 m by 0.3 m by 0.25 m and may be the core of a wall where the facing stones had been robbed and re-used. The survival of masonry and foundation at this depth would suggest that this may be the remains of a wall or structure that formed part of the monastic precinct rather than the later tenement buildings.

To the south of the central baulk layer 012 was excavated to reveal a similar deposit consisting of an extensive mixed layer of dark grey/brown silt (014) up to 0.8 m thick with frequent inclusions of mortar, slate, animal bone and late 18th or 19th century artefacts including pottery and glass. Below this was another layer of similar material (019) up to 0.4 m thick with occasional fragments of charcoal as well as brick and tile. Before the baulk was removed as thin lens of orange/red silty sand (022) was visible in section and this may be the southern extent of deposit 013 although this lens dipped to a much lower level possibly tipping down from the masonry 025. Once the baulk was removed and the western sections cleaned it became apparent that there was a layer between 014 and 019 identified by a much higher concentration of light brown mortar (037) but otherwise very similar. This layer appeared to be tipping down to the south again possibly from wall/masonry 025. At the limit of excavation of layer 019 next to the masonry (025) there was a small patch of orange/red burnt material (026) approximately 0.25 m in diameter and 0.04 m thick, no artefacts were present but the material was sampled for analysis. As all these layers contained high concentrations of mortar it is likely that they represent demolition layers.

As the excavation proceeded more of the structural remains of the medieval drain were exposed, the large flat mortared stones (020) that were uncovered below the concrete on further investigation were revealed to form the main body of the roof and north wall of the drain. Excavation of the loose sand (011) uncovered the facing stones (028) of the north wall. Two courses of squared stones the largest up to 0.3 m by 0.3 were uncovered and appeared to be partly built over a deposit of mid-brown silt (023) that had been cut by the sand deposit 011, a medieval pottery strap handle from a jug, a fragment of slag, possibly furnace lining and several fragments of tile were recovered from 023. Excavation of 023 revealed further courses of masonry (030) 0.2 m below the first. Again two courses were exposed but this time the stones were smaller (0.2 m by 0.1 m). The deposit 023 changed too and became lighter in colour with more mortar fragments (021). Layers 023 and 021 appear to be contemporary with the construction of the drain.

In the south east corner of the trench excavation revealed a section of the in situ fabric of the medieval drain where the facing stones and part of the rubble core had been removed. At this point it was possible to investigate the relationship between deposit 023 and the core of the wall. The deposit 023 extended into the core of the wall and narrowed to a thickness of only 3-4 centimetres. Therefore the gap in the facing stones filled by 023 may be the result of subsidence and partial collapse of part of the drain or it may reflect different phases of construction, certainly the previous investigations inside the drain has demonstrated different phases of building.

In the south east corner closest to the intact medieval drain the deposits were slightly different in that they were tipping to the west as well as to the south. Layer 014 sealed 019 but also partly sealed a layer of loose brown mortar with angular stone rubble (021) that extended 1.2 m west into the trench and was up to 0.2 m thick. This in turn sealed a layer of mortar rich rubble (036) that was tipping down to the west where the stones in the rubble were larger up to 0.2 m by 0.15 m. As the excavation was finishing a small slot trench 0.4 m wide was excavated against the south baulk at the deepest point of the excavation to further investigate deposit 036. At this point there were far fewer stones and the deposit consisted of mainly light brown mortar and sand along with several flat stones (039). These deposits relate to the demolition of the drain at this point as there was no surviving in situ masonry. Unfortunately it was not possible to fully investigate these deposits.

As the excavation was coming to an end and the final cleaning and recording was underway several angular stones were uncovered within a reddish brown silty clay (040) in the narrow strip between the wall of the drain 030 and the masonry foundation 025/029 between deposits 019 and 024. The wall of the drain was stepping out as the excavation got deeper it may be that this material was associated with the foundations of both structures and if continued they would have met approximately 0.8 m to the south.

On completion of the excavation the deeper sections of the trench and the possible medieval features were protected with geotextile prior to backfilling. The trench was backfilled with the material that had been excavated and stored next to the trench. The Parks department then re-laid the turf

Artefacts
Most of the artefacts recovered from the excavation date to the late 18th or 19th centuries with very few medieval or late medieval pottery sherds. Generally the medieval pottery sherds were recovered alongside 15 later material from mixed deposits although the deposits uncovered at the end of the excavation did not appear to be disturbed (contexts 019, 023, and 024). What was unusual about the artefacts from these deposits was the large amount of roof tile

This limited excavation confirmed the presence of deep deposits of demolition material and landscaping as was expected particularly to the west of the manhole where the drain was known to be blocked. What was not expected was the survival of walls relating to the recent tenements that occupied part of the site in the 19th century as earlier attempts to locate these walls had failed and it was presumed that these foundations had been grubbed up during the demolition and landscaping of the site in the early 20th century. More importantly the excavation revealed the masonry foundations of a wall or building to the north of the drain that may be contemporary with the drain and relate to the monastic precinct.

Parts of the exterior of the monastic medieval drain were also uncovered and this was the first time that these had been exposed and recorded. Although the interior of the drain adjacent to the excavation trench consists of a well made pointed arch the exterior of the drain had a flat top and slightly stepped vertical sides where they were partially uncovered on the north and west side, suggesting that the exterior drain formed a square masonry block. The outside face of the wall to the north had two distinct types of facing stone with a silt deposit separating them suggesting two periods of building or possibly of repair. To the west of the manhole where the drain is known to have been blocked the upper surface of the drain was missing to reveal the mortared masonry core that would have formed the thickness of the wall and arch. Due to the reduced area due to the depth of the excavation the trench was probably located over the thickness of the wall rather than fully over the blocked arch at the mouth of the drain certainly the deposits in this area appeared to be tipping to the south and west and consisted of compacted mortar rich silts with a few small angular stones. Apart from part of the outer face and masonry core uncovered closest to the manhole no medieval structural remains were uncovered although the deepest deposits (036 & 039) did contain more frequent and larger stones and rubble but nothing that could be described as building stone for an arch or interior facing stones. As the location of the trench was over the fabric of the wall rather than the inner wall face mortared rubble for the wall core would have been more likely. As the floor of the drain is approximately 3.9 m below ground level the deepest point in the excavation would still have been approximately 1.5 m above floor level. At this level if the drain had continued as a pointed arch 2.2 m high and 1.9 m wide (taken from RCAHMS survey) it may have been expected to find walls surviving or collapsed stones from them the fact that no large faced/dressed stones were recovered at this depth would suggest that the drain had been demolished facing stones had been robbed for re-use in the same way that the masonry (025) had no facing stones.
Text from GUARD DSR