||Fieldwork: Summer Excavation 2001
Interim Report by Stephen Young
The existence of a Romano-British settlement at Whitehall Farm was initially established in 1996 after chance finds of Roman coins, pottery and building material were made by Mr Dave Derby and Mr Steve Pulley whilst metal detecting. Further to these discoveries and with the agreement of the landowner, extensive fieldwalking, geophysical investigation, preliminary trial-trenching and limited area excavation were undertaken between the years 1996/2000. The artifactual and structural remains indicated an extensive area of occupation consistent with a Roman villa complex. It also revealed the existence of an ancient agricultural landscape covering a core area of at least five hectares. The diagnostically datable material recovered suggested active occupation of the area from the late Iron Age until the end of the Roman period and possibly even into the early Anglo-Saxon epoch.
Unfortunately extensive destruction of the archaeological stratification has occurred across key structural elements of the site. These disturbances are mainly due to the Victorian robbing and quarrying of building material from the site and the subsequent degradation caused by post medieval agricultural practice. The main villa range is particularly effected in this respect with the surviving elements being severely truncated and denuded. However the associated bath house, field systems and trackways have survived to a far better degree and offer improved prospect of archaeological investigation.
The Whitehall Farm Villa (SP6495860) is typical of the extensive Roman settlement located in this part of Northamptonshire. It forms part of a broader network of sites situated between the Roman small towns of Towcester (Lactoduorum), Whilton Lodge (Bannaventa) and Duston. The current series of excavations offers an excellent opportunity to examine the impact of Romanisation on this outpost of empire.
The second phase of archaeological exploration of the site began last year (2000) with the commencement of large-scale excavation of the site.
This year's work has focused on four different areas of interest. We have continued to develop our understanding of the overall villa plan by uncovering more of the main villa range with particular emphasise on trying to understand the layout of the east wing.
A suite of three rooms associated with the hypocaust (Caldarium) and stoke room (Praefurnium) of the bath house partially uncovered in 1999/2000 have also been investigated. In addition a third building initially identified last year about 100ms to the south west of the villa has been explored. The final area of archaeological interest was a trench cut through a bank and ditch that formed an enclosure around the structural elements of the villa.
The Villa: Main Range and North Eastern wing
The absence of coherent structural detail for most of villa building has meant that interpretation of the remains is not as straight forward as it could possibly have been. Residual surface finds indicate that 3rd and 4th century foundation layers survive but that they are heavily disturbed and need to be examined in greater detail to elucidate the finer points of the architectural plan and chronology. However it is possible to confirm that the building is a winged corridor villa of modest size and pretension. The central range appears to have been constructed in the mid to late 2nd century and presumably was subsequently embellished with the addition of mosaics and a fronting corridor with additional wing pavilions during the second half of the 4th century.
The villa was built along a north east and south west axis and faced south east. The archaeological evidence suggests a two storied stone structure (30m x 10m) which could have contained between twenty or thirty rooms. At the eastern end of the main range was an apsidal/semi-circluar shaped room. The position of several of the robber trenches for the corridor and wing pavilions have been located.
The excavation of this structure revealed a rectangular stone building (15m x 7m) containing five rooms aligned west to east and terraced into the hillside. It appears to have been initially constructed in the mid to late 2nd century and to have extensive alterations to two rooms situated at the extreme west and east ends of the building in the late 4th century. This year the excavation revealed an additional three rooms which are likely to be associated with the apodyterium, Frigidarium and Tepidarium. These rooms would have been designed respectively for dressing/undressing, a cold room for cooling down towards the end of the bathing experience and a warm room as an intermediate acclimatising room before entering the hot room (Caldarium) found in last years excavation. Indeed the extensive alteration to Room 4 suggests that the mostly easterly end of the bath house was added or entirely rebuilt in the late 4th century. The overall plan demonstrates that the building is a typical example of an estate bath house and would have been capable of meeting the full range of bathing requirements.
In Room 3b as part of the collapsed floor fill we found the moulded base of a plunge pool which because of its proximity to the hypocaust (under floor heating system) is more likely to be associated with activities located in the hot room. Indeed the blocked east/west flue alignment of the hypocaust appears to suggest that the original caldarium also included this room in its initial layout. Virtually all of the Roman floor levels have disappeared although in one place the cement base for the tiled floor can still be seen. The continued retrieval of painted plaster from this area should also enable us to reconstruct elements of the interior design in this part of the bath house.
The bath house enjoyed a long working life which continued into the very late 4th and possible early 5th century. Its eventual demise was due to the slow deterioration and collapse of the fabric as witnessed by the fallen stone roof slates impacted into the ground surface outside the northern external wall of the bath house and the evidence of a fallen interior connecting archway between Room 3b to 4. In the last days of the excavation we encountered exciting new features below the floor level of Room 4 which include two culverts and at least one holding tank. These monumental structural elements represent the survival of an earlier bath house construction. Although the area is as yet little understood it would appear to contain a unique range of water management systems whose interpretation could well explain important aspects of day to day water utilisation in the bath house during the Roman period.
Proto-villa / ancillary farm structure
A metalled floor surface discovered in 2000 was investigated further in the hope of recovering its plan and some evidence of its chronology. At first it appeared that the floor belonged to a building which might have been contemporary with the development of the main villa range. The building could have provided accommodation for estate workers or have been used as a storage barn. It was also possible that the structure might have been an example of an aisled barn; a common building type that has often been identified further east in the Lower Nene Valley.
After preliminary examination and interpretation of the excavated evidence it can be demonstrated that the building is domestic in nature and can best be described as a proto-villa or strip house. These buildings or long houses always have a series of three or four interlocking rooms with access from an external veranda.
The construction of the building revealed it had no stone walls and that the metalled surface in fact represented the surviving floor levels. The building would have been rectangular (10m x 5m) most likely making use of vernacular building methods and materials but in a Romanised style. Its superstructure was probably based on large wooden beams laid on the ground surface into which posts and timber cross bracing would have been inserted while wattle and daub would have been utilised to fill any holes. The exterior of the building might well have been rendered and the roof would have been thatched. No internal partitions were recognised during the excavation although inside the structure was a hearth (10m x 5m).
The surprising thing about the structure is its chronology and relationship to the main villa range. The strip house is not contemporary with the villa and indeed seems to precede its construction. Although 2nd/3rd century residual pottery was discovered associated with the metalled surface the feature itself had been cut into by three pits with pottery of the mid to late 2nd century suggesting the floor of the strip house could not have been laid after that date. Therefore the building must have been constructed before that date most likely at the end of the first or early 2nd century. The destruction or demolition of the building at a relatively early date is further supported by the insertion of two ditches into the metalled surface of the strip house floor during the 3rd or 4th century. In fact the larger of the two ditches is aligned along the long axis of the strip house and may well be part of the enclosure ditch to the villa complex. One other feature located in this area which is of interest and requires further investigation is a post pit which may be evidence of a veranda.
Villa enclosure bank and ditch
A section was excavated across a possible ditch with an east/west alignment to the north west of the villa range, which had been previously identified by the geophysical survey. We hoped to ascertain whether the feature was a perimeter barrier enclosing the villa. The nearby Horestone Brook Roman villa certainly had an enclosure delineated by a bank and ditch and it would not be unreasonable to expect such a feature to be a diagnostic characteristic of villa classification in the area.
The excavated section revealed the depth of the ditch to be 2 metres and the width to be over 4 metres. In addition on the south side of the ditch was a bank which survived to height of 1 metre. The alignment and location of the bank and ditch strongly suggest the existence of an enclosure surrounding the villa. Unfortunately no dating evidence was recovered from the fill of the ditch and therefore it is impossible to prove the hypotheses beyond all doubt at this stage. The underlying reasons for constructing a bank and ditched enclosure around the principal buildings was not necessarily defensive but to indicate clearly the division between domestic and agricultural aspects of the estate and also to restrict the access of wild animals. It should be remembered that bear, boar and wolf were still relatively common in the Roman period.
The excavation has been largely funded by a grant provided by the Friends of the Upper Nene a community based forum supported by South Northants District Council. We would also like to thank Nick and Roz Adams the owners of Whitehall Farm for their permission to excavate the site and their continuing friendly interest and unstinting effort in organising this year's season of work. Sponsorship and professional/technical assistance from Newmodernmasters.com and oliomedia has been much appreciated, as has the huge amount of effort and commitment demonstrated by the many local volunteers organised by Kate Widdall. In addition we are grateful for the support of companies like Lloyd Fraser Supply Chain and B&Q whose generosity has eased our operational costs and burdens. All of the people helping on site are volunteers, many are local although we have some individuals from places around the West and East Midlands and Yorkshire. The core of the "digging team" is made up of students from University College Northampton and other universities like Birmingham, Exeter, Leicester, Sheffield, Glasgow and UCL.