ARCHAEO-BLOG - A PROJECT DIARY OF THE ARCHAEOLOGICAL SURVEY AND EXCAVATION IN DINEFWR PARK, LLANDEILO, CARMARTHENSHIRE, 2008

	THE DAVID AND CHRISTOPHER LEWIS FOUNDATION

During the spring of summer of 2008 Dyfed Archaeological Trust in conjunction with the National Trust will be carrying out several excavations and surveys in Dinefwr Park. The David and Christopher Lewis Foundation, the National Trust and the Cambrian Archaeological Association are generously funding the work.

Dinefwr Park is noted for its picturesque landscape. But it is the numerous archaeological sites within the park that make it unique. These include: two Roman forts; the ruined medieval castle of Dinefwr; the deserted medieval town of Dinefwr; the deserted medieval town of Newton; 17th-18th century formal gardens around the Newton House and numerous other remains.

Several elements of work are planned: topographic survey around Dinefwr Castle and elsewhere in the park, geophysical survey and test-pitting and trial excavations on the site of Dinefwr town, test-pitting on the site of medieval town of Newton; and trial excavation on the site of a possible Roman building within the Deer Park.

This work follows on from a successful community excavation on the Roman forts in 2005. For the dig diary of this project , click here.

Volunteers working on the Roman fort excavation in 2005

VOLUNTEERING OPPORTUNITIES AND OPEN DAYS

Opportunities are available for members of the community and other interested people to participate in the excavations or join guided tours of work in progress. All the archaeological work will be supervised by professional archaeologists.

23 - 28 June 2008. Volunteers are welcome to participate in trial excavations around Newton House on the site of the medieval town of Newton. The 17th-18th century formal gardens were also located here.

30 June to 14 July 2008. Volunteers are welcome to participate in trial excavations on the site of a possible Roman building in the deer park.

Saturday 28 June 2008. An open day, with guided tours around the excavation around Newton House and demonstrations of the survey and geophysical equipment. Normal admission charges apply.

Saturday 12 July 2008. An open day, with guided tours around the excavation on the possible Roman building and demonstrations of the survey and geophysical equipment. Normal admission charges apply.

ARCHAEO-BLOG

March - May - topographic survey of the deserted town of Dinefwr

It was hoped that detailed topographic survey in woodland around Dinefwr Castle on the possible location of the deserted medieval town would be completed before vegetation sprang into life. We were almost successful – the last few days of the survey were in hot weather when vegetation was bursting into life. Owing to the dense woodland it has never been possible to properly interpret the earthworks that may relate to the town. The emerging survey is already helping in the interpretation.

19th - 20th May

During the course of two warm sunny days several members of Dyfed Archaeology Trust excavated 23 1m square test pits in woodland around Dinefwr Castle in the anticipation of locating the site of Dinefwr medieval town. It was hoped that artefacts such as pottery present in the topsoil would provide a clue to the town’s location. However, no medieval artefacts were discovered (the only finds were a sherd of Roman pottery and a prehistoric flint flake), and there was no charcoal, bits of burnt bone and other debris one would expect on a long-lived settlement. This was very surprising, as some of the pits were located close to the castle. It is possible that town’s inhabitant’s had few pots and other implements to lose, but it seems more likely that the town was not situated immediately outside the castle, but more distant from it. The search will continue.

Towards the end of the topographic survey in May when spring burst in to life surveying conditions were not ideal. Fortunately, from May onwards most surveying was done in the non-wooded areas of the park.

May – June 2008

Survey and excavation on the site of Newton.

Although no traces are visible today, medieval documents tell us there was once a town split between two locations in what are now the grounds of Dinefwr Park. One, Dinefwr, sometimes called ‘Old Town’ or Upper Town’ was primarily a Welsh settlement established around Dinefwr Castle probably during the 13th century when the castle was under Welsh control.

The other, ‘New Town’ – Newton – sometimes called ‘Lower Town’, was established in the early years of the early 14th century in order to help secure English control of the area. The last record of Newton appears in the 16th century, when it was described as “a street now ruinous”. The first mansion or house of Newton was was built within the ruins of the town. It is likely that all surface traces of the town were erased when extensive gardens were laid out around the house in the 17th century.

A geophysical survey, in conjunction with a topographic survey, undertaken in late May – early June 2008 detected extensive buried remains, some of which related probably relate to the medieval town and others to the formal gardens.

Geophysical survey. The strong ‘dotted lines’ are modern pipes and cables, and the rectangular area of high interference to the north of the house is an old tennis court. A series of rectilinear enclosures can be seen to the north and west of Newton House. Two alignments are visible, one possibly relating to the town and the other to the gardens. To the east of the house the buried archaeology seems less complex, but two distinct but partial rectangular enclosures are visible.

Phil Poucher undertaking the topographic survey on the site of Newton

23 June 2008

Searching for Newton – the first day of excavation. Four trenches were opened.

Trench 1 was positioned over a rectangular enclosure to the east of the house. A least two parallel ditches were immediately revealed. Excavation on these started, but as yet no dating evidence has been found.

In Trench 2 excavation began of appears to be the 4m – 5m wide ditch a rectangular enclosure to the south east of the house. The size of the ditch was surprising, and perhaps indicates a Roman or prehistoric site rather than medieval. Little in the way of dating evidence was discovered, but a sherd of Roman pottery was found in soil above the ditch.

Lengths of what are probably ditches of property boundaries of the medieval town were discovered in Trench 3 to the west of the house. But this interpretation is not certain - further excavation will reveal more.

A mass of rubble mixed with 18th century and earlier pottery, glass and other finds were found in Trench 4 to the southwest of the house. This material seems to be associated with the remains of two walls. These may be part of the formal gardens.

24 - 25 June 2008

Work continued on cleaning and excavating the four trenches. Archaeological deposits are present in all four trenches, but as yet no evidence for the town of Newton has been found.

In Trench 1 a whole series of parallel ditches were revealed, the largest of which contained much iron slag, but unfortunately as yet no dating evidence has been revealed. The ditch in Trench 2 is beginning to be as big as suspected, with a depth so far approaching 1.5m. A sherd of Roman pottery is the only indication so far for its date.

The ditches in Trench 3 are reluctant to reveal much about themselves. No dating evidence has yet been discovered and therefore their assignment to the medieval period, as previously thought, is in doubt.

The remains in Trench 4 in contrast are both complex and rich in finds. Foundations of walls have been found, with artefacts in and around them suggesting they were mostly removed in the 18th century. They are quite substantial, indicating something more than garden walls, but their function and date has to be established.

26 - 27 June 2008

Two wet days. Despite the weather progress was made. Recording of Trenches 1, 2 and 3 was completed. It has not been possible to come to firm conclusions regarding the remains in Trench 1, but it seems most likely that they are not medieval but Roman, as is the large ditch in Trench 2. This ditch may be part of a practice camp, associated with the nearby Roman fort.

Further analysis and on the ditches in Trench 3 indicates a medieval date. They are possible boundary ditches associated with the medieval town. The largest ditch may even have had a defensive function.

The walls in Trench 4 are probably of a building. Their date is unclear, but they may be of a late medieval or early modern house rather than walls of the 18th century formal gardens. A pit alongside one of the walls is getting deeper and may be an in-filled well.

28 June 2008

An open day was held for members of the public to view the excavations and see the finds. Guided tours were provided and demonstrations of geophysical equipment given. Approximately 75 people joined the tours and a similar number made casual visits.

Apart for final recording, the only excavation took place in Trench 4. The archaeological remains in this trench are more complex that previously thought, and what may be the remains of medieval timber buildings have been revealed pre-dating the formal garden.

The finds have been washed, allowing for a more consideration of the pottery. What was initially thought to be Roman pottery from the large ditch in Trench 2 is in fact medieval. It is possible that this ditch is therefore part of a moated site, perhaps surrounding the original, 15th century, Newton House. However, it will not be possible to demonstrate this as most of this site lies beneath the National Trust car park and mature trees.

Monday 30 June is the last day of excavation on the trenches around Newton House. From Monday excavation will move to the site of a possible Roman building at the Heronry Dam.

Excavation of the site of a possible Roman building – June - July 2008

Geophysical survey on the site of a possible Roman Building. This site lies on the western side of the deer park, close to the Heronry dam – the pond is now dry. Earthworks of an old road, probably the medieval route from Newton west to Carmarthen lie here. Roman pottery and building material found in the stream below the dam strongly suggests a Roman building in the vicinity – possibly a bath house associated with the forts, or a temple. The geophysical survey shows complex buried remains including a possible Roman road, an enclosure around the pond and other ditches and enclosures. These will be investigated in June and July.

30 June 2008

Work started on opening five trenches on the site of a possible Roman building. Apart from Trench 9, which was located over a the site of small 19th century building, and which contained a mass of building material, little archaeological was immediately visible. However, most trenches contained Roman tile and brick and occasional sherds of Roman pottery – rare artefacts in west Wales. More careful cleaning of Trench 5 revealed a spread of rounded boulders – these require more investigation.

1 July 2008

Trench 6 had been positioned over a geophysical anomaly and a slight linear earthwork – possibly the line of a Roman road. No evidence for this is present. The earthwork seems to be composed of topsoil. Excavation is continuing. In contrast, the spread of stones running across Trench 5 has the appearance of a Roman road, but again further excavation is required. Cleaning of Trench 7 has begun. No excavation has yet been undertaken on Trench 8. The building remains in Trench 9 are beginning to resolve themselves into a well preserved cobbled floor and mortared walls.

2 July 2008

A warm day punctuated by heavy showers. The remains of the post-medieval building in Trench 9 are more extensive than expected, with pitched-stone floors and walls occupying the whole area. These remains are so well preserved that they will be cleaned, photographed and planned, but not removed.

The spread of boulders in Trench 5 is looking more and more like a road. If it is a Roman road then it is in an unexpected position, and running to and from unpredicted locations.

Work continued in Trench 6 revealing more Roman pottery, including a sherd of Samian Ware, and a piece of glass, but no structural remains are present.

Further geophysical survey was undertaken, this time in an area of an open valley immediately to the east of the excavation. The survey has revealed little of interest.

3 July 2008
Work progressed despite the very wet weather. A new trench was opened (Trench 10) at the side of the old pond. Initial results are promising, with Roman pottery and building rubble present. Final cleaning took place in other trenches and planning started.

4 July 2008

Dyfed Archaeological Trust trustees visited the site following a management committee meeting. There was some debate over the nature of the archaeology. In particular they considered the ‘Roman road’ in Trench 5 unconvincing. They favoured an interpretation for it as a foundation for a massive bank. Jeff Davies thought the pottery covered a wide time range, from the 2nd to 4th centuries. The site was not, therefore, associated with the nearby Roman fort, as this came to an end in the first half of the 2nd century.

In Trench 8 a ditch associated with an usual ‘ladder’ shaped geophysical anomaly is rock-cut, narrow and deep. It contained two sherds of Roman pottery, but its function is presently unclear.

7 July 2008

After a weekend of rain the first part of the morning was taken up by bailing out the trenches. Following this exercise it rained heavily for the rest of the morning!

A section across the stone spread in Trench 5 was started, and extended into a bank at the east of the trench. This bank seems to be part of the agricultural landscape pre-dating the creation of the deer park in c.1660.

The results in Trench 10 are not as promising as first thought. The rubble and Roman pottery and brick seem to be above a clay deposit, which is possibly the clay lining of the 18th century Heronry Pond.

Planning of the post-medieval remains in Trench 9 started, whilst excavation continued in Trench 8. Two parallel ditches in Trench 8 seem to be part of a trackway. Only Roman pottery has been found in these ditches.

8 July 2008

A wet morning, again, followed by a sunny afternoon. In Trench 8 there are now three parallel ditches, of varying sizes, but all V-shaped, with Roman pottery the only finds. The function of these is still unclear, but two seem to be the ditches of a ‘ditched trackway’ and the third part of the square enclosures visible on the geophysical survey.

Work continued in cutting a section through the stone spread and bank in Trench 5. The date of these features is still unclear as no artefacts have been discovered in association with them. A narrow gully in the base of Trench 6 is possibly prehistoric – several worked flints were found in and around it.

Further excavation in Trench 10 confirmed that the clay layer is 18th century pond lining. All the Roman finds – brick, tile, pottery and glass are above this layer.

No definite evidence for a Roman building in the form of walls or floors has yet been discovered, but the building materials and other finds clearly indicate the presence of one in the immediate vicinity of the Heronry Pond. A hand-dug trench away from the pond revealed no trace of a building. More hand-dug trenches will be excavated over remaining few days of the dig.

9 July 2008

Heavy rain all day. There was no work on site.

10 July 2008
Most of the morning was spent bailing out the trenches following the heavy rain of the previous day. This took up most of the morning. A new trench was opened in the field to the east of Trench 5 over a geophysical anomaly. No trace of structures was found, and there were no artefacts.

11 July 2008
The last full day of excavation. Mostly a day of recording, although excavation continued in Trench 7 and a small new trench was opened upslope and to the east of 7. There was nothing in this new trench. More Roman brick and tile continued to come from 7, and it looked increasingly likely that the deep deposits in this trench are sitting in a large ditch or terrace rather than a natural hollow.

12 July 2008
The last day – an open day. Approximately 45 people attended two guided tours of the excavations – each tour lasting almost two hours. Final recording was completed, and excavation in Trench 7 continued. This trench continually filled with water making excavation difficult. However, right at the end of the day a layer of Roman brick and tile was uncovered at the bottom of the trench, possibly a building demolition layer. This layer and associated deposits continued below the bottom of the trench, but could not be examined owing to its depth, ground water and lack of time. A considerable amount of earth moving will be required if this deposit is to be examined in the future.

In summary, the presence of large amounts of brick and tile indicate a substantial Roman building in the area of the excavation trenches, but it could not be pinned down. Pottery suggests that the building was in use from the 1st to the 3rd/4th centuries and so is unlikely to be solely of military use. On the final day of the excavation evidence from Trench 7 indicates that the building may be sealed by over 1.5m of silts and soil.

[click for navigation menu if not present]