1990 GGAT Archaeological Assessment — Land at Great House Farm, Llandough — Transcription

Transcription note: this is a transcription of the supplied scan. The PDF is image-based and contains no usable text layer, so OCR has been used. Obvious OCR uncertainty is retained or marked where the scan cannot be read reliably. Figures, plans and maps are identified separately rather than being silently converted into text.

Title pages

LAND AT GREAT HOUSE FARM, LLANDOUGH
AN ARCHAEOLOGICAL ASSESSMENT
The Glamorgan-Gwent Archaeological Trust Ltd.
August 1990

LAND AT GREAT HOUSE FARM, LLANDOUGH
AN ASSESSMENT OF THE ARCHAEOLOGICAL IMPLICATIONS OF REDEVELOPMENT AT THE SITE OF GREAT HOUSE FARM, LLANDOUGH
The Glamorgan-Gwent Archaeological Trust Ltd.
August 1990

1. Introduction and acknowledgements

Following the demolition of Great House Farm, Llandough in December 1988 the Glamorgan-Gwent Archaeological Trust (GGAT) entered into discussion with Debenham Tewson, agents for the site owners B P Properties Ltd, on the possible archaeological implications of redevelopment. The Trust, in its capacity as archaeological advisor to the Vale of Glamorgan Borough Council (VGBC), informed the local planning authority of the archaeological sensitivity of the site, and also kept in close contact with the Llandough Community Council, who initially contacted the Trust about the redevelopment.

It was agreed that the potential archaeological interest should be considered in the light of Welsh Office planning guidance and the County and Local Plans. Accordingly the VGBC determined that, in granting outline planning consent, “no development shall take place until a field evaluation survey and archaeological assessment report has been carried out by a qualified archaeologist to the satisfaction of the Local Planning Authority” and that the detail planned shall pay full attention to the results and findings of the assessment.

A [short] month time-limit was set for the archaeological work. On the recommendation of Morgan Bruce, and Debenham Tewson, it was agreed that the report should be completed as soon as possible thereafter.

B P Properties commissioned GGAT to undertake the work of which this introduction forms a part. It is evident therefore that GGAT has throughout played a dual role: on the one hand it has advised the Local Planning Authority, and on the other provided comparable advice to the applicant. It should be stressed that GGAT made this clear to both parties at the outset, in order that either of them could seek alternative qualified advice if it was felt that a conflict of interest might arise.

The assistance of a number of individuals and organisations is gratefully acknowledged by GGAT. In particular Mr Joy of Debenham Tewson, Messrs Dylan and Maitland Evans of VGBC, Mr Cann of Morgan Bruce and Mr Egan of Llandough Community Council all afforded every co-operation.

2. The redevelopment site

The redevelopment site occupies the area to the north and east of St Dochdwy’s church. The boundaries are demarcated by the break in slope above Llandough Hill to the north, by adjacent housing to the east, and by the churchyard and the northern area of the development site. [Some text on this page is unclear in the scan/OCR.]

Figure 1: Great House Farm, Llandough. Location of re-development site.

3. Archaeological background and potential

The church of St Dochdwy and the surrounding area is of considerable archaeological interest. The churchyard is likely to be on the site of an earlier foundation, perhaps dating back to the seventh century AD (Davies 1978). The cross shaft to the south of the west end of the church dates to the tenth or eleventh century and is of a scale and style which could mark an important religious site (RCAHMW 1976, Inv. No. 828).

To the south of the church an Iron Age farmstead, Roman [occupation], medieval monastic grange, late medieval [village/site] and farm are known to have existed. There is also the presence of a Norman earthwork castle. [Several words on the supplied scan are indistinct.]

For the purposes of this assessment the principal development itself was Great House Farm. [The report discusses] an ownership dispute and the subsequent demolition in 1988. The building was [of] considerable archaeological interest; although both structural and documentary evidence was incomplete, the farmhouse may have contained parts of a seventeenth-century house.

The seventeenth-century phase is associated with the Vaughan Family, the nineteenth-century work with the Bute Estate. A number of architectural features were recorded in situ in 1974 (Fig. 2) and others were observed in the rubble following demolition. Sherds of twelfth- to fourteenth-century pottery were recovered by E. J. Beare in 1962-3 on the slope immediately north-east of Great House, while it is likely that some medieval ironwork which is held by the National Museum of Wales was discovered under the rear wing c.1880 (RCAHMW 1988).

In addition to the recorded evidence noted above there was potential for further archaeological interest. The possibility that deposits relating to the early monastic settlement might survive in this area could not be ruled out, and it was possible that the churchyard may formerly have been more extensive. Although the main focus of Iron Age and Roman activity appeared to lie to the south it was possible that further evidence might have survived in this area. A medieval well, Ffynnon y Drindod, was originally located to the north of the church, and although its exact position is unknown it may have been located within the development site (RCAHMW 1988).

4. Field objectives and methodology

The aim of the archaeological fieldwork was two-fold. The first instance the investigation was aimed at establishing whether deposits relating to any of the features described in 3 above (or indeed any other type of archaeological activity) were present. The second was to assess the likely impact of redevelopment on the identified archaeological interests and to recommend to B P Properties methods by which these interests could best be protected. In addition to the fieldwork other relevant sources (e.g. cartographic) would be consulted as appropriate.

In view of the size and character of the development it was decided that the most appropriate way to test the archaeological potential of the site was by a series of machine-cut trenches, the position of which was determined either by the likelihood of archaeological deposits being encountered (e.g. near the churchyard wall), or at random. It was agreed that the trenches should mostly be cut to the level of the substrate, but in some areas this proved impractical. The machine trenches were closely observed and subject to basic archaeological recording.

The system of machine trenching was in effect a compromise between a more extensive assessment of the archaeological potential of the site, and the pre-existing state of knowledge, which in this case was insufficient to enable informed archaeological decisions to be taken. The results therefore give a much more reliable indication of the likely archaeological implications of redevelopment, yet by their nature cannot provide total assurance on the presence and value of archaeological deposits across the development site as a whole.

5. Archaeological fieldwork results

The approximate location of the machine-cut trenches is shown on Figure 3, and the trench numbers given below also correspond with Figure 3.

It was originally intended that a test pit should be dug manually in the south-east sector of the site where machine access was problematic, but it was felt that the stability of the higher ground to the north, in the area of “The Gardens”, was questionable and that such an excavation would be ill-advised.

The build-up of demolition material in the vicinity of “The Gardens” also led to a modification of the original plan to cut a trench adjacent to the north part of the east wall of the churchyard. In the event this cut (6) was made nearby at the east end of the north wall of the churchyard. The results from trenches 1 and 5 suggested that the other proposed excavations in the vicinity of these trenches were unnecessary. With the above exceptions the investigation was undertaken on the basis of the agreed sketch plan which formed part of the contract between the Trust and B P Properties Ltd.

Trench 1

This was approximately north-west/south-east across the line of a depression, shown on the OS map. The ground had clearly been disturbed by the recent levelling operations.

For most of its length the trench cut through a layer of soil containing considerable quantities of gravel, brick and stone, to an average depth of 1m. This overlay the clay substrate.

Towards the south-east end of the trench its character changed. The top c.0.2m-c.0.3m consisted of soil and stone and brick rubble, which overlay c.1m depth of grey/brown soil with brick-dust, bricks and stone rubble. Below this the clay substrate was cut by a large feature aligned north-east/south-west, in excess of 1m in depth. The north side was angled at about 45° while the southern edge of the feature sloped at a much shallower angle. The feature was over 8m wide and extended beyond the southern end of the trench and below the depth to which the trench could be cut safely. Its fill consisted of very dark grey clay and a considerable quantity of steel wire.

In the southern part of this feature c.0.6m below its top, there was a lens, up to 0.15m thick, of brown clay with charcoal flecks, brick fragments and brick-dust. This layer extended beyond the south-eastern end of the excavated trench.

The principal feature was almost certainly the depression indicated on the OS map, which was formerly a quarry road. The depth of the road confirmed that it would have cut any earlier archaeological features which may have existed. The absence of any archaeological deposits between the substrate and the modern material in the north-west part of the trench suggests that this area is likely to be sterile. Although based on a small sample we therefore believe that the area to the north-west of and including the quarry road is unlikely to contain significant archaeological interest.

Trench 2

Trench 2 was cut on an approximate north-south alignment and excavated to a depth of up to 2m. The lowest 0.5m of this depth was probably the clay substrate within which what appeared to be a thin bank of lias bedrock was observed. This deposit was overlain by c.0.5m of a grey clayey soil containing bricks and lumps of mortar. The top 1m of material was formed by a sequence of deposits containing modern debris including plastic, and a thin skim of topsoil.

As in Trench 1 the absence of any archaeological layers between the substrate and comparatively recent deposits indicates low archaeological potential in the vicinity of Trench 2, thus extending the probable archaeologically sterile zone to the south of the quarry road identified in Trench 1.

Trench 3

Trench 3 took the form of a T with the long arm cut on a north-south alignment, and the short arm extended to the west. These excavations were located in the vicinity of the farm outbuildings shown on the OS map.

The north-south trench revealed a sequence of post-medieval and modern deposits of loose clayey soil with brick, charcoal and mortar. At the north end of the trench these deposits covered the clay substrate to a depth of up to 2m, but towards the south end there was a marked increase in the depth of overburden. Here the deposits became significantly wetter, with a high proportion of redeposited clay and lias, all of which appeared to be loose infill. The limits of this infill were unclear nor could its depth be recorded as excavation was halted for safety reasons at a depth of c.3.5m below the ground surface.

In the western arm of the trench the deposits described above overlay the clay substrate to a depth of c.1.5m, but in certain areas intervening deposits were recorded. On the south side of the cut c.2m from its junction with the north-south trench a lens of brown soil overlying the substrate was found to contain fragments of human skull and long bones. At the west end of the trench a thicker deposit (c.0.2m) of brown clayey soil was recorded; this contained large quantities of charcoal and some animal bone. On the north side of the cut c.4m from the trench junction an ill-defined feature was sealed by the post-medieval deposit, although it appeared to be cut into identical material at a depth of c.1m below the present ground surface. The feature was c.0.2m deep by c.0.4m-0.5m wide and was filled with, inter alia, hard sandy mortar similar to that used in the area during the medieval period.

The post-medieval clay and rubble deposit was defined in the western arm by the foundations of a north-south post-medieval wall which abutted the layer. This wall and a parallel example c.6.5m to the west doubtless represent the footings of the barn which stood on the site until 1988. Some traces of the mortar floor of the structure survived. Between the two wall footings the place of the clay and rubble deposit was taken by a dark grey-brown rather clayey soil sealed beneath the barn floor. The barn floor, walls, etc. were overlain by c.0.2m of disturbed topsoil etc., evidently redeposited during the demolition process.

Within narrow machine-cut trenches there must be some doubt over the exact interpretation of the sequence and features described above. The two areas where an archaeological deposit was noted between the substrate and the more recent material clearly have an archaeological interest. The human bone appeared to be disarticulated and part of the general deposit, indicating that it may have been disturbed from its original context. These layers are discussed in greater detail below together with the associated evidence from Trench 4. The depth of post-medieval infill in the southern part of the north-south trench is suggestive of a quarry pit, other examples of which are known in the area. It is likely that this was deliberately infilled in preparation for the construction of the farm outbuildings, with a general spread of rubble etc. being spread across what was to be the yard area, subsequently defined by structures such as the barn, the footings of which were identified in the western trench.

In summary the area which is presumed to have been quarried has low archaeological potential, but elsewhere it appears as if archaeological deposits pre-dating the construction of the farm buildings, presumably in the 19th century at the behest of the Bute Estate, have survived.

Trench 4

This trench ran along the northern edge of the tarmac-surfaced roadway at the southern limit of the site parallel to the adjacent churchyard wall.

The clay substrate was encountered at a maximum depth of more than 1m; throughout most of the trench the substrate was 1.1m below the present ground surface.

Below the tarmac, which varied in thickness from c.0.1m to 0.2m, was a varying depth of grey-brown soil mixed with stone and brick, which appeared to be levelling material for the roadway. Between this layer and the clay substrate was a layer of grey-brown clayey soil up to c.0.2m thick. This contained numerous human bones. These were scattered along most of the length of the trench. Most of the bones appeared to be disarticulated and in places bones from several different parts of the skeleton were found close together. For example the mandible and base of a skull were close to a femur, humerus, carpals, and lumbar vertebrae with an axial vertebra in close proximity. It is likely therefore that the bones had been disturbed from their original context. There appeared to be a general tendency for the bones to be aligned approximately north-south. There was no evidence of associated graves.

As in the western extension of Trench 3 there is clear evidence for the survival of archaeological deposits which predate the recent activity on the site. The fact that the substrate rises towards the present ground surface to the north further implies that it has been truncated by the construction of Great House Farm itself. If the latter was initially of medieval date it may be that the deposit containing human bone was derived from the levelling of a former graveyard associated with the church. Alternatively it is possible that the human bone was disturbed in situ by other activity. While these and other possibilities remain difficult to determine without a direct stratigraphic link to Great House, the likelihood is that the ossiferous layer and the early deposits overlying the substrate are of medieval or earlier date.

Trenches 5–8

Trench 5: a short approximately north-south aligned cut originally intended to traverse the nineteenth-century structure attached to the Great House. In the event it proved more practicable to place the cut across the front boundary wall of the Great House. Approximately 0.5m depth of predominantly nineteenth-century rubble directly overlay the clay substrate. If any deposits, such as those recorded between the substrate and more recent layers in Trenches 3 and 4, had existed in this area they had clearly been removed by nineteenth-century activity.

Trench 6: located adjacent to the east end of the north section of the churchyard wall and aligned approximately east-west. The upper deposits were similar to those described under Trench 4 and had formed part of the drive in front of the farm. An underlying deposit of clayey soil and rubble rested directly on the surface of the substrate at a depth of c.1.1m below the present ground surface. If any earlier archaeological deposits had existed here they had been removed by more recent activity.

Trench 7: intended to be located on an east-west alignment immediately to the north of Great House but a minor discrepancy resulted in the trench clipping the north part of the former house. The north-east corner of the house, probably the dairy, was cut into the clay and lias limestone substrate with a stone floor, probably laid in the eighteenth century, at a depth of c.1.0m below the present ground surface. The north-south partition wall of the dairy was mortared and 0.36m wide. The trench cut obliquely along part of the north wall of the house, so neither the plan view nor the section of the south side of the trench was entirely clear. Pots of paint and a plastic pipe within this basement area were overlain by mortared rubble demolition presumably dating to 1988. The entire structure had evidently been built within a construction pit. The pit and demolished structure were overlain by redeposited topsoil with mixed rubble etc. to a depth of c.0.3m.

To the west of the dairy no trace of the structure of Great House was identified, with c.0.3m of topsoil as described above directly on the clay substrate. The trench did not extend sufficiently far to the west to determine whether or not the foundations of the west wall of the farmhouse survived. To the east of the dairy once again the mixed topsoil directly overlay the substrate.

It was evident from this excavation that only features cut into the clay and lias limestone substrate had survived. Any floor levels associated with the north-west part of Great House Farm must have been removed during demolition in 1988 or during the clearance of the demolition in 1989.

Trench 8: aligned approximately east-west near the crest of the scarp above the former Brickworks, and to the north-east of the site of Great House Farm. Approximately 0.5m of mixed topsoil, rubble etc. directly overlay the substrate, suggesting that little of archaeological significance has survived in the immediate surrounding area.

6. Interpretation of results

The results of the trial excavations suggest that various parts of the redevelopment site have differing archaeological potential. For the purposes of this report the archaeological potential has been grouped under four headings:

1. Non Existent – Low Archaeological Potential
2. Medium to High Archaeological Potential
3. Unknown Potential but estimated to be Non Existent to Low
4. Unknown Potential but estimated to be Medium to High

The parts of the site which display these characteristics are outlined below and illustrated in Figure 4, which forms the basis of the recommendations.

Low Archaeological Potential: Trenches 2, 5, 6 and 8, and the layers located between the east and west arms of trench 3 and the north-south element of trench 3 were similarly sterile. The east-west arm of trench 3, all of trench 4, the southern projection and the central part of trench 7 all contained deposits of potential value. By taking account of positive and negative archaeological evidence it is possible to suggest a zone of non-existent to low archaeological potential which extends across most of the northern area of the site as shown on Figure 4.

Medium to High Archaeological Potential: Trenches 3 and 4 both contained up to 0.2m of archaeological deposits, which inter alia contained human bone, charcoal, etc. These layers must be regarded as of high archaeological potential on present evidence, although it should be noted that further excavation could either enhance or lower this perceived potential. By combining both positive and negative evidence as outlined above it is therefore possible to define a zone of medium to high archaeological potential to the north of the yard as shown on Figure 4.

The cutting of trench 7 identified part of the north-east corner of Great House Farm, which had survived demolition as its substructure below ground level at this point. Although of limited extent, the remains have a surviving archaeological potential.

7. Unknown potential and recommendations

Unknown Potential but estimated to be Non Existant to Low Potential.

For reasons described above much of the eastern area of the redevelopment site could not be investigated. The combination of the data from the other trenches, taken together with surface and cartographic indicators permits an informed estimate of the likely archaeological potential to be made.

It is likely that the absence of archaeological deposits on the northern part of the site, as demonstrated by trenches 7 and 8, is repeated further to the east. It is also likely that in the area of “The Gardens”, a demolished nineteenth-century property, earlier archaeological deposits will have been disturbed or removed. To the south of “The Gardens” a steep slope drops to the valley bottom. Although the latter cannot be dismissed entirely, all this area is regarded as low in estimated archaeological potential.

Unknown Potential but estimated to be Medium to High Potential.

In view of the archaeological deposits in trenches 3 and 4, other areas adjacent to the churchyard wall may have potential. The land between the churchyard and the former site of “The Gardens” is one such area. There is also a possibility that the deposits recorded in the west arm of trench 3 extended further to the north, south and west. These areas are therefore regarded as high in estimated archaeological potential.

Archaeology and redevelopment: options and recommendations for further action.

The information obtained from archaeological fieldwork suggests that a large part of the site has a low archaeological potential, but that redevelopment is constrained by factors of archaeological interest in other parts of the site where suitable recording measures will be desirable.

It is recommended that those areas marked as of non-existent potential (both actual and estimated) should be the subject of a low-intensity watching brief. This exercise would act as a failsafe to record features cut into the substrate, e.g. pits, the Ffynnon y Drindod well which might have escaped detection during the fieldwork. It is also recommended that the surviving parts of Great House Farm, e.g. the north-east corner, are recorded by means of a watching brief during development.

Those areas, both actual and estimated, of medium to high potential present a different problem. The objective should always be to protect archaeological interest. The deposits in the west arm of trench 3 were shallow enough that it may be possible to protect them from damage by ensuring that foundations are not cut to a greater depth than the deposits. This may depend on the load-bearing capacity of the site and the type of houses proposed. An alternative, bearing in mind the proximity of the pit in the north-south arm of trench 3, would be the layout of the development so as to make this area over to garden space.

In the part of the site surrounding the west arm of trench 3 the report recommends two options: (1) protection of the archaeological interest by minimisation of ground disturbance; or (2) if this is not possible, excavation and recording of the archaeological layers prior to development. Option (1) is preferred if at all possible.

The archaeological deposits identified in trench 4 were only shallowly buried. Even if the area were set aside as garden space within the development design there would be a risk of deposits, including human skeletal remains, being disturbed by routine gardening activities. The only effective way of protecting the archaeological interest in this area would be to prevent any further ground disturbance and raise the ground level by c.0.5m, an operation which might present problems for the churchyard wall. If this raising of ground level is impractical there will be little alternative to the excavation of the archaeological deposits in advance of development. The report therefore recommends two options for the area of and surrounding trench 4: (1) protection of the archaeological interest by preventing further ground disturbance and raising the ground level by at least 0.5m; or (2) if this is not possible, excavation and recording of the archaeological layers prior to development. Option (1) is preferred if at all possible.

In both the areas referred to above (i.e. in and around trenches 3 and 4) there may be an additional legal complication. If the presence of human bone is taken to denote the existence of a disused burial ground the provisions of the Disused Burial Grounds (Amendment) Act 1981 may apply. Section 2(1)(a) of this Act requires the removal and re-interment of human remains prior to the erection of buildings, although section 2(2) does make provision for the Secretary of State to over-rule this requirement if he is satisfied that the remains will not be disturbed. The report therefore recommends that before any decision on the options given above is taken, advice should be sought by B P Properties Ltd on any legal requirements which may pertain to the human skeletal remains.

Finally there is the potentially sensitive area to the east of the churchyard. Although the archaeological interest is not proven in this area it is possible that similar layers to those identified in trench 4 may be present. As they would be off the line of the driveway of Great House Farm it is likely that any such deposits will be protected by a greater depth of overburden, but this likelihood cannot be guaranteed. The protection of the possible archaeological interest could be achieved by the spreading of the rubble from “The Gardens” over the sensitive area preparatory to the laying of garden topsoil. Such an operation might cause problems for the churchyard wall. Two options for further action are therefore presented: (1) protection of the possible archaeological interest by the prevention of ground disturbance and the raising of the ground level; or (2) if (1) is not possible, any archaeological layers which may be present should be excavated and recorded prior to development.

Figures in the supplied report

• Figure 1: Great House Farm, Llandough. Location of re-development site.
• Figure 2: plan/drawing of Great House Farm and recorded architectural features (supplied scan).
• Figure 3: Great House Farm, Llandough. Approximate position of trial excavations and principal archaeological features.
• Figure 4: Great House Farm, Llandough. Archaeological Potential Zone.

Bibliography / abbreviations

Davies, W. 1976. An Early Welsh Microcosm: Studies in the Llandaff Charters.

Owen-John, H. S. 1988. “Llandough: the Rescue Excavation of a Multi-Period Site near Cardiff, South Glamorgan” in Robinson, D. M. (ed.), Biglis, Caldicot and Llandough: Three Late Iron Age and Romano-British Sites in South East Wales, Excavations 1977-79. British Archaeological Reports 188.

RCAHMW, 1976. An Inventory of the Ancient Monuments in Glamorgan, Volume I: Pre-Norman, Part I: The Early Christian Period. Royal Commission on the Ancient and Historical Monuments in Wales.

RCAHMW, 1988. Unpublished records relating to Great House Farm prepared by R. F. Suggett, December 1988.

SMR 20382. Sites and Monuments Record No. 2038, South Glamorgan. (Record held by GGAT.)

Thomas, H. 1988. “Historical Notes” in Owen-John 1988.