A new project at Butser Ancient Farm
Fifty years ago the Council for British Archaeology set up Butser Ancient Farm to explore life in Iron Age Britain. Its latest project is a Neolithic house. Rachel Bingham, Trevor Creighton and Claire Walton, Butser staff, and Gareth Chaffey, the structure’s excavator, describe the thinking and effort behind the UK’s newest ancient building
Developer-led archaeology rattles along at a 21st-century pace, with ancient landscapes gobbled up by housing estates and infrastructure projects almost as quickly as they are exposed. Developer-funded initiatives offer tantalising glimpses of the ancient wonders uncovered, through exhibitions, school visits and perhaps an article or two in the press or British Archaeology, before the landscape is once again transformed in the next chapter of our human impact.
Rarely does an archaeologist have the reward of seeing the fruits of their excavations brought to life in a tangible structure – pits and postholes, stratigraphy and soil samples turned into something you can walk through, and touch, feel and smell, long after the diggers, bulldozers and builders have removed all traces at the site. A collaboration between Butser Ancient Farm and Wessex Archaeology is looking to offer just that. We are creating a building that will offer a unique, immersive journey into an imaginary world of our remote predecessors, in the process shedding light on Neolithic construction techniques. The story begins, as it often does in archaeology, at a quarry.
First find your house
Wessex Archaeology has been working at Kingsmead Quarry, Horton, Berkshire for cemex uk since 2003. A large area has been excavated and recorded on the northern floodplain of the River Thames, just west of
Heathrow Airport. It’s long been a favourable place for settlement, for which there is copious evidence, from hunters and gatherers at the end of the Ice Age (Britain in archaeology Mar/Apr 2018/159) to medieval farmers.
Between 2008 and 2012 Gareth Chaffey and his team discovered four Early Neolithic (4000–3000bc) structures at Horton (News Sep/Oct 2008/102). All four were rectangular in plan, but built to two different designs: two were post-built, relatively small and with limited artefacts and dating evidence; and two had clear foundation trenches, with signs that these had held upright wooden posts and planks.
The artefacts found at each of these four structures suggest that they were used, at least for some of their lifespans, as houses: these include distinctive Early Neolithic pottery, burnt flint, flint tools and manufacturing waste, along with fragments of animal bone, charcoal and charred plant remains such as cereal grains and hazelnut shells. Radiocarbon dating revealed that all the buildings were largely contemporary, perhaps used over two or three generations some 5,700 years ago.
Neolithic houses are rare in the uk, and to find four on one site is even more so. During the analysis, Wessex archaeologists long discussed how these structures may have looked or how they were built. They created illustrative reconstructions to visualise possible design, engineering and architectural solutions, proposing walls built of split oak planks roofed with timber or turfs. The archaeologists had no practical experience of raising such structures, and the design was guided by the evidence from the ground as far as was possible. The models provided a sense of completion and resolution to the discovery. After a flurry of press interest, public site openings and report writing, the dig team moved on to further projects, the Horton site taking its place in the literature and in the memories of all those who excavated there.
Last summer Wessex Archaeology received an email that turned that on its head. Butser Ancient Farm was advancing plans to replace its existing Stone Age building, which had been based on evidence from excavation of a Neolithic house at Llandygai, North Wales. It was not long before they had settled on Horton as a perfect archaeological source. Butser had learnt about the Horton excavations online, and appeared to find a likeminded group in Wessex Archaeology, equally as excited and engaged by the (perhaps daunting) prospect of bringing the archaeology to life.
After several meetings and discussions, it was agreed to make an interpretation of the biggest of the four houses from Horton. The building (referred to as Horton 2) measured around 15m long and 7.5m wide. Its form was irregular: it tapered towards the eastern end, and both ends bowed inwards slightly. A substantial post stood in each corner, as well as two internally to create a partition which divided the space into two separate rooms. There was no evidence for floor surfaces or a hearth, but a possible entranceway for the building had been identified.
The carefully recorded evidence and the scale of the house fitted the bill in showing the potential size and sophistication of Early Neolithic construction. These features also suited Butser’s needs, in providing a practical educational space for the
growing number of school groups, who – in normal times – visit the site each year. Horton is also a relatively local site to Butser in south-east England, helping to tell the story of some of the earliest settlement and habitation in the area. Now it was a case of translating this evidence into the tangible, practically testing out the theories and designs speculated by Wessex Archaeology through Butser’s speciality: experimental archaeology.
Unexplored possibilities
The idea of testing theories about the past through practical experiments is not a new one. John Coles coined the term “experimental archaeology” in the early 1970s to describe, as he saw it, “any honest effort to understand ancient artefacts by actually working with them”.
Lest this bring to mind visions of re-enactors wielding swords, or industrious metalworkers and craftspeople beavering away in period costume, exponents of experimental archaeology are determined that it be seen as a process of scientific research. As Dennis Harding, an archaeologist with a special interest in Iron Age houses, put it, “The basic objectives of scientific experimental archaeology are first, to test archaeological interpretations by empirical experiment and replication, and second, to conduct such tests in a way that permits feedback into excavation practice, data retrieval and analysis.”
One of the best known practitioners of this approach was Peter Reynolds, the first director of Butser Ancient Farm. The project was founded in 1970 by the Council for British Archaeology, recruiting Reynolds to establish a working “ancient farm” on Butser Hill near Petersfield, Hampshire, where archaeologists could experiment with their theories on how people lived in Iron Age times. Work started on a trial site known as Little Butser in 1972, with the first public open day in 1974. A second site, known as the Ancient Farm Demonstration Area, was opened on nearby Hillscombe Down in 1976. In 1991, with both these earlier sites closed, the project moved to Bascomb Copse on the slopes of Windmill Hill, between Chalton and Clanfield, 5km from the original Little Butser.
With his stalwart volunteers, over nearly 20 years Reynolds gathered data on soil preparations, sowing, growing, harvesting and storing grain in pits, and rearing animals. However, it was his construction of roundhouses at Butser, in particular the Pimperne House, that really captured people’s imaginations.
Basing the entire structure on nothing more than the visible traces and postholes that formed the archaeological footprint of a large Iron Age roundhouse in Dorset, where the reports had included features that left the excavators scratching their heads, Reynolds’ construction revealed a possibility hitherto unexplored by traditional theoretical methods.
The farm is now close to celebrating its 50th year of experimental work. Reynolds’ legacy continues apace at Butser, with a number of experimental buildings spanning over 5,000 years from Neolithic to Anglo-Saxon times. In each case, the techniques, exploration of materials, consumption of resources and erection of the building, are only the start of the experimental process. In the longer
term archaeologists at the site monitor the durability of the chosen materials and thus the potential lifespan of such constructions, feeding back to the wider archaeological community to inform interpretations and identification on the ground.
With the evidence recorded by Wessex Archaeology at Horton, and the considerable experience of experimental construction from the team at Butser, it was only a few months before theoretical discussions turned into practical actions. All talk of stratigraphy and excavation became that of structural engineering.
A ten ton hogback roof
For a structure to be truly experimental, we have to consider the practicalities of which resources, tools and technologies are known to have been available to the original builders. Structural elements in Horton 2 have been fastened using simple lapped notches, drilled and secured with oak pegs, and lashed with cordage. Techniques were tested using replica Neolithic tools including flint axe blades, bone drills and timber mallets, and the structure erected entirely with human labour. In this way the team was able to validate archaeologically their design and construction methodologies. It is also important to note that the capabilities of the team in using Neolithic tool types is undeniably significantly below the best standards of antiquity!
The excavated footprint of Horton 2 presented a significant design challenge; the archaeology revealed just six posts, only two of which offered internal support to the overall structure. This is an astoundingly minimalistic structure by any standards, and certainly in stark contrast with the more generous sprinkling of postholes found in most continental and Irish Neolithic houses, where they are more common than in England.
The only certain attribute of Horton 2 was its trapezoidal floor plan. For the project team headed by Butser archaeologist Claire Walton and project coordinator Trevor Creighton it was the commitment to adhere to this which informed many assumptions shaping the building’s final form. First and most significant was that it had a roof. The foundation trench, punctuated by probable opposed posts at the corners and mid-point, invited interpretation as a post-in-trench, interrupted sill-beam structure, with unknown intermediate walling. Internally, only two postholes were visible, and it was this combination of evidence and assumption that produced the striking load-bearing
a- frame design of the house we built, with earth-fast posts at each corner and at the mid-point of either wall.
The proximity of the original site to extensive wetlands and rivers makes water-reed thatch plausible as the roofing material. Considerable thatching experience has shown that the principal rafters would therefore need to extend upwards from each post at an angle of 45–50 degrees – the necessary pitch for thatch waterproofing – meeting centrally at a height of around 5m. In turn this configuration essentially necessitates tie beams to avoid lateral collapse: the rafters would need to be linked at the bottom, or the weight of the thatch would flatten the roof.
With thatch weighing ten tons and the building spanning over 7m, modern health and safety and engineering precautions have necessarily also had some influence on the design. Two intermediate roof trusses – not specifically indicated archaeologically but also not counter indicated – were included to strengthen the structure. This also influenced our decision not to have load-bearing side walls. These
would have required a substantial and continuous wall plate on the top to support the roof, as well as leading to an extraordinarily tall structure to maintain the required pitch. By using earth-fast principal rafters, the ground acts as tie beam and wall plate together, creating a much more secure structure.
Most strikingly, through following the ground plan faithfully and maintaining constant pitch, the roof exhibits a curving, profoundly “hogback” profile. While this is implicit in the ground plan, the dramatic effect in the built form emphasises this structurally enhancing attribute, perhaps once common among the many prehistoric buildings of trapezoidal plan.
Our walls are not structural, but represent three possible interpretations of the original – planking at the east end, wattling on a sill plate at the west, and low, lateral weather walls on either side. These are important variations archaeologically – their residue can be analysed after demolition to see how the evidence fits with the original archaeology, giving the house a significant legacy long beyond the lifespan of the structure itself. They are also significant educationally – highlighting the fact that experimental reconstruction is always the selection of options from many possibilities.
Enriching validation
The house we are creating at Butser is not, of course, the only possible interpretation of the Horton 2 structure, but that does not invalidate it experimentally. Horton 2’s largely unsupported 7m vault is unlike more common, post-built Neolithic structures, bringing novel challenges.
The measures required to overcome these and best ensure structural integrity are often overlooked in theoretical reconstructions; only through experimental construction can we begin to understand how these structures may have worked. The minimal detail seen at the Horton site in fact allows for a greater number of possible interpretations than those excavated houses with more postholes and features. Wessex’s original plankwalled interpretation looks quite different to Butser’s “roof on the
ground” solution – and there are other possibilities.
Experimental archaeology is a wonderful tool for showing us what could have happened, rather than what did happen. The number of possible interpretations of the Horton house based on the same archaeological footprint bears testament to this fundamental truth. That said, far from just “playing with mud and stones”, experimental archaeology is arguably a more scientific approach to interpreting the past than widely accepted academic methodologies that formulate visions through theoretical discourse alone.
“Plausible” is the highest level of certainty in experimental archaeology. The Butser Horton house is unquestionably a plausible interpretation, already yielding experimental dividends. It is also a potent representation of the scale and ambition of Neolithic builders.
By adding to the dating and excavation evidence from Horton, the Butser construction has an important role to play in discussions about the structure and form of our earliest permanent settlements and homes. Alongside its unique experimental interpretation, the substantial structure offers potential insight into questions of temporality, long term investment in a place, resource requirements and use over several generations. Analysis of the building’s functionality and longevity at Butser will continue to add to this fascinating picture.
What of the impact of the project on those involved? For Gareth Chaffey, and the many other field archaeologists from Wessex Archaeology who have had a hand in the experimental construction, the process has already had a profound effect. Building a house has had an immediate impact on their approach to excavation and interpretation. For the team at Butser Ancient Farm, working so closely with the very people who excavated the structure has been a hugely enriching process. Access to wider archaeological expertise, specialisms and support, together with the profile of Wessex Archaeology, has enabled this latest construction to delve much deeper and reach a far wider audience than would have been possible on their own. Conversations between excavators and experimental archaeologists have raised understanding and appreciation of Horton 2. This is a great validation of experimental archaeology, and the mutually beneficial approach of commercial and experimental archaeologists working so closely together. It is to be hoped this will be the first of many such projects at Butser Ancient Farm.
At the time of going to press, Butser Ancient Farm is closed and most staff have been furloughed in line with government guidelines in relation to covid-19. The house remains unfinished; construction progress to date is detailed in blogs at www.butserancientfarm.co.uk/blog and www.wessexarch.co.uk/our-work/ butser-project-building-neolithichouse. Rachel Bingham is Butser Ancient Farm creative developer; Trevor Creighton is Butser Ancient Farm project coordinator; Claire Walton is Butser Ancient Farm archaeologist; Gareth Chaffey is a senior project manager at Wessex Archaeology