Horton’s  Neolithic houses

13 mins read

Exploring  a prehistoric landscape at  Kingsmead  Quarry

Over a decade of excavation in a quarry near Horton, Berkshire, has laid bare a remarkable prehistoric site, boasting some of the finest Early Neolithic buildings in the country. Alistair Barclay and Gareth Chaffey described to Matthew Symonds how the arrival of farming transformed a landscape.

The remains of four, or perhaps five, Early Neolithic houses have been found within Kingsmead Quarry. This was the first example. Discovered in 2008, it measures 9.8m by 6.5m, and was divided into two rooms by the partition walls visible either side of the scale.

Passengers gazing out of the window as their flight climbs west away from London Heathrow are initially greeted by the familiar sight of the M25. Beyond this ugly tarmac sliver, often partly obscured by the equally familiar sight of queuing traffic, lie the gleaming expanses of the monumental Wraysbury and The Queen Mother reservoirs. But a keen observer will make out, midway between these eye-catching examples of post-war engineering, a narrow ribbon of yellow, where the topsoil has been stripped away, incised for almost a kilometre across the landscape. Overlooked, literally, by millions of passengers every year, catching sight of this innocuous extraction site made one observer grab for his camera. That was because the Kingsmead Quarry at Horton is one of the most remarkable prehistoric sites in England, and the passenger, jetting off on his honeymoon, was Wessex Archaeology’s Site Director of excavations there, Gareth Chaffey.

Gareth’s photograph now serves to illustrate the quarry location, but in many ways the scale of Horton’s archaeology is best appreciated from above. Wessex Archaeology has been excavating the Kingsmead site for CEMEX UK since 2003, during which time almost 34 hectares have been investigated. As the topsoil is machined away prior to gravel extraction, the entirety of the underlying archaeology is laid bare. What can, at ground level, appear to the casual observer to be little more than a disorientating mass of linear bands criss-crossing the natural brickearth becomes recognisable from a higher vantage as  a gigantic palimpsest landscape. Here, the sinuous course of long-silted palaeochannels and the boundaries of prehistoric fields and settlements become distinct.

Lying just west of London Heathrow, and the Wraysbury reservoir, the CEMEX UK works at Horton are overlooked by millions of passengers annually. On this occasion, the passenger was Site Director Gareth Chaffey, en route to his honeymoon.

This superimposed set of features compresses over 12,000 years of human history, and contains Neolithic buildings and a barrow, Early Bronze Age ring ditches and a Beaker burial, two Middle Bronze Age farmsteads, an Iron Age settlement, and a Romano-British farmstead. Together they testify to the gradual evolution of a managed landscape in what was once a pocket of land bound within a Thames meander, at a time when the river flowed 3km north of its present course. In particular, Horton’s archaeology provides an opportunity to examine how natural woodland where hunter-gatherers roamed was ordered into farmsteads and field systems.

Foragers and flint-knappers

The oldest find from Horton is far earlier than any of the ditches inscribed into the brickearth. It is a cordate hand-axe manufactured around 300,000 years ago. This flint implement was found, not by the archaeologists, but by one of the quarry workers who plucked it from the gravel that was being whisked away for processing on a conveyor belt. Well suited to delivering a cutting and  chopping action, this style of axe is often associated with Neanderthal craftsmanship, suggesting that modern humans were not the first hominins to work this landscape. Clusters of Late Upper Palaeolithic flint have also been found at the site, and, although all of these deposits appear to have been disturbed, the presence of knapping detritus in the form of flakes and cores as well as blades makes it likely that tools were being manufactured in the area around 12,000 years ago.

The totality of the archaeology detected by Wessex Archaeology at the Kingsmead Quarry site. Covering almost 34 hectares, manmade features span the period from the Early Neolithic through to the post-Medieval period, although the greatest activity occurs between the Neolithic and Roman periods. Among the features, the four Early Neolithic houses (circled blue), and a Middle Bronze Age foundation deposit (circled green) are the most remarkable. Most of the field boundaries dividing up the landscape date to the Middle Bronze Age, although those directly to the right of the post-Medieval oval feature in the northern wedge of excavations belong to the Roman period.

The earliest archaeology surviving in situ belongs to the Mesolithic period, and consists of a handful of flints discarded in a shallow hollow. This modest discovery provides a fleeting but vivid glimpse of everyday life for hunter-gatherers operating in the middle Thames valley. The hollow was formed when a scoop of earth was gouged out by a root system when a tree came crashing down. The natural worksurface this created caught the eye of a passing hunter, who took the opportunity to augment their tool kit by knapping a few extra blades. After discarding the flint core, and a scatter of flakes and blades, the visitor went on their way once more. The general scarcity of Mesolithic flints from across the site makes it unlikely that hunter-gatherer bands regularly camped there.

‘You can’t put forward any suggestion of continuity from the Late Mesolithic period into the Early Neolithic, as there’s virtually no evidence for Late Mesolithic activity,’ Alistair Barclay, post-excavation manager and one of the authors of the forthcoming excavation report, stresses. ‘So, from that point of view, we favour the pioneer-farmer model of the Early Neolithic at Horton. This is a time when there was renewed contact with mainland Europe. It is worth  bearing in mind that there is Late Mesolithic activity in the region, but it is occurring a few kilometres up-river. Of course, the Mesolithic can be difficult to identify, but the main thing is that it’s not happening at Kingsmead!’

It is the wealth of Neolithic archaeology discovered at Horton that makes it an exceptional site. Among the various artefacts and structures signalling the dawn of farming in this country, Neolithic houses are one of the earliest to appear. Most date to around 3800-3600 BC, pre-dating the earliest causewayed enclosures by a good century or so. Ranging in length from a modest 6m to over 20m in Scotland, most of these houses were built of timber, leaving little more than a cluster of post-holes to mark their presence. Phenomenally hard to spot unless large areas are stripped, the growth of developer-driven archaeology has seen a surge in their numbers. Over two dozen Early Neolithic houses are now  known in but Britain, but they are still rare.

‘It’s very unusual to find more than one example on a site,’ says Gareth Chaffey. ‘At Horton, we have at least four Neolithic houses, which is exceptional. We also have a possible fifth, which we refer to as a “house void”. This is a cluster of pits, possibly formed when the house timbers were dug out for re-use, which mirrors the size and shape of another house about 30m away. The Kingsmead site gives us a fantastic opportunity to learn about Neolithic settlement in the first few centuries of the 4th millennium BC.’ The evidence from Horton is particularly valuable because not only does it furnish important structural details, it also adds to debate raging about the role of these buildings in the new social order sweeping through Britain. A particular area of contention is whether they were halls where dispersed communities could gather, or domestic dwellings inhabited by families.

This Early Neolithic house was discovered by the Wessex team in 2012. Measuring 15m by 17.5m, the outline of its walls is displayed by the course of foundation gullies. Leaf-shaped arrowhead and small fragments of plain bowl ware were found in association with the house.

First farmsteads

The four certain examples of Neolithic houses from Horton display two different construction styles. Two of the houses survive as no more than groups of post-holes, while the other pair have gully foundations that show both the course of the external wall, and the presence of an internal partition dividing both buildings into two rooms. ‘All four houses are roughly based around six posts,’ Gareth explains. ‘The first house we found, in 2008, covered about 9.8m by 6.5m, and was  built using gully construction. In one area, we found traces of possible wooden planks set into the foundation trench. It was quite a finds-rich structure, and contained broken pottery, worked flint, charred plant seeds, animal bone fragments, polished bone points, and a nice fragment of a Langdale axe that was manufactured in the Lake District. In 2012, we found a structure that was about a third larger, measuring 15m by 7.5m, but otherwise very similar. Together these represent two of the best-preserved Early Neolithic  structures discovered in this country.’

Intriguingly, the scale of the gully foundations  increases at the same point in both buildings. In each case, the slots lining the eastern cell were two or three times deeper than those to the west, leading the archaeologists to speculate that the eastern room may have supported a mezzanine floor. Indeed, reconstructions of the houses — based primarily on the 2008 example — include a ladder leading up to a possible storage  or sleeping space. This detail was based on the discovery of a pair of suitably proportioned post-holes that had been driven deeply into the ground near the partition wall.

The four certain examples of Neolithic houses discovered at Horton. Top are the two gully-built houses divided into two rooms by partition walls. In both cases, the foundations are deeper to the east, perhaps indicating an upper floor. Below are the two post-hole houses. The one to the right was found in 2011, and closely parallels a Neolithic house found almost 250km away in Derbyshire.

To get an idea of what the other upstanding architectural features may have looked like, the team turned to the surviving timbers from the Sweet Track, a raised walkway traversing boggy ground in the Somerset Levels, constructed in 3807-3806 BC. ‘The joints in Karen Nichols’ reconstruction mimic the ones found in the Sweet Track,’ Alistair reveals. ‘And it does seem that part of the trackway could have been built using timbers that were originally intended for a house.’

Although the archaeologists drew inspiration from a trackway, it is possible that the houses themselves were instrumental in influencing the design of a very different style of monument. Both the 2008 and 2012 houses have curving end walls that bow inwards in an unusual and distinctive way. These evoke the curving façades that are sometimes displayed by long barrows. As these repositories for the ancestors are an approximately contemporary development, it is possible that the architecture of the dead was deliberately referencing that of the living.

Among the second group of houses — those represented purely by post-holes — is a building that cribbed from more profane source material. Discovered in 2011, it consisted of 23 post-holes set around four deeper and stouter uprights, arranged in an off-square configuration at the heart of the building. While these internal posts must have formed the load-bearing frame and carried the weight of the roof, almost like a square version of the later Bronze- and Iron-Age roundhouses, the truly remarkable feature of this design is that it is a near carbon-copy of a structure known as Lismore Fields building 1, discovered 250km away in Derbyshire (CA 103). Superimposing the two house-plans shows an astonishing degree of overlap. ‘You can only build houses this similar if you have seen them,’ Alistair says. ‘It is beyond coincidence, so there has to be some connection between the builders of the  2009 house at Horton and Lismore Fields. We don’t know how this knowledge was exchanged, but it must have been shared in some way.’

A reconstruction of the house found in 2008. It is based on both structural traces found at the site, and material incorporated in the broadly contemporary Sweet Track in Somerset. In the background, a new structure is being erected — the so-called ‘house void’. Inside the 2008 house (right), a mezzanine floor is shown over the eastern room, complete with a ladder based on the location of two post-holes found within the structure.


Public or private?

Unfortunately, in all four — or perhaps five — cases at Horton, only the house foundations survived and the Neolithic floor surfaces had been destroyed by later activity. This did not prevent the archaeologists from reaching some firm conclusions about what the houses were used for, though. One of the strongest clues came from the nature of the artefacts in the houses, and where they were found. ‘We sampled 100% of the remains of these structures,’ says Alistair, ‘and when you look at the finds, like the small pieces of pottery trapped in the gullies, it looks exactly like people were cleaning out the buildings and sweeping things against the walls. So I think we’re seeing domestic activities being carried out by people living within them.’

This post-built Early Neolithic house was discovered in 2011, and covers 7.7m by 5.5m. Displaying a different construction-technique to the other Horton houses, with internal, load-bearing timbers, it is a near carbon-copy of a building erected at Lismore Fields in Derbyshire. This raises intriguing questions about how knowledge was exchanged in the Neolithic.

Magnetic Susceptibility survey conducted over the 2012 house by the Wessex Archaeology team  points to a similar conclusion. The technique is designed to detect traces of enrichment caused by human activity, and the archaeologists hoped it would pinpoint the location of the hearth. But, when they excavated, the area it highlighted proved to be a doorway. ‘So, despite not getting the hearth material or a floor surface or anything like that, we did  manage to pick up the inhabitants’ sweepings out of the door!’, Gareth explains. It is a tiny detail that is incredibly evocative of the chores performed by house-proud home-owners living  in the Thames valley almost 6,000 years ago.

The exceptional density of Neolithic houses at Horton raises the question of whether these structures were isolated buildings that were rebuilt on new plots once they decayed, or the residences of a dispersed community living and working in the area at around the same time. Only the 2008 house returned enough radiocarbon dates to allow its origins to be refined by Bayesian modelling. Charred cereal grain and hazelnut shells from the building suggest construction within the period 3710-3650 BC. Both the radiocarbon dates and the pottery from the adjacent house void suggest a slightly later date, perhaps equivalent to a gap of a couple of generations. This prompted Karen Nichols to devise a reconstruction showing the 2008 house gone  to seed, while in the background a new structure is being erected on the site of the house void.

This plan shows where finds were recovered within the 2008 house. The quantity of small artefact fragments discovered in the foundation gullies has been taken by the Wessex team to be a consequence of domestic sweeping. Among these objects were a piece of a Langdale axe and plain bowl pottery sherds. These tantalising relics of the detritus of everyday life hint that the houses were lived in, rather than serving as halls for a dispersed community.

There are certainly hints that the different houses were constructed by people aware of the alignment, size, and location of neighbouring properties. ‘I think that we do have evidence for a settlement here,’ says Alistair. ‘We are not looking at isolated buildings. This is not a unique arrangement — after all, Lismore Fields in Derbyshire produced three buildings. But it does beg the question of why we do not find more of these buildings elsewhere. At Kingsmead, finding so many buildings was partly down to luck, but it was also important  to get the excavation strategy right. Of course none of this would have been possible without the generous support of CEMEX UK.’

Beaker burial

After this remarkable Early Neolithic surge in  activity, there is a lull until the Middle Bronze Age. It is, though, to the Early Bronze Age that another of Horton’s extraordinary discoveries belongs. It was no more than an innocuous blemish in the brickearth when it was first spotted, and the archaeologists tasked with investigating the feature assumed that it was  another void torn out of the ground by a tumbling tree. Excavation, however, told a different story. Within lay the poorly preserved remains of an individual over 35 years old who had been laid to rest with their head facing south. Although the bones were too badly degraded to be able to determine sex, the orientation of the head suggests that the deceased was probably female. If so, the skeleton’s sex only serves to make the accompanying grave goods even more remarkable.

The interment belongs to the group known as ‘Beaker burials’. As well as the eponymous pot, the ‘lady’ was laid to rest with an exotic set of beads crafted from amber, lignite, and some of the earliest gold ever found in Britain. These hint at connections to a much wider world, with the lignite coming from East Anglia, the gold possibly from Cornwall or western Britain, and the Baltic amber probably gathered on the shore of the east coast of England. The gold beads are made of sheet metal, some of them displaying perforations, thread marks and possible decorative lines, perhaps suggesting that they had been cut from other objects and recycled. Regardless of their origin, there is no question that this individual must have been an important member of her community, perhaps a spiritual or political leader.

During the Middle Bronze Age, a remarkable deposition was made in an oven pit or hearth deposit. Carefully arranged in a circle at the base of the pit (left) these objects date from the Late Upper Palaeolithic through to the Middle Bronze Age. Spanning thousands of years, whether these artefacts (right) were curated heirlooms or casual finds cannot be said, but they clearly play with notions of ‘the past’.

It is in the Middle Bronze Age that Horton’s landscape underwent its most  dramatic transformation. For the first time, it was carved up into substantial and extensive field systems. Two farmsteads dating to this period were established within the excavated area, but the most intriguing discovery was a far more modest one. A remarkable group of artefacts were found carefully placed in a circle at the base of a heavily charred oven pit or hearth deposit. These objects included eight barbed and tanged arrowheads of different styles, two flint scrapers, a broken flint blade, a whetstone, and a bronze awl.

Together these objects span an impressive timescale. The bronze awl would best fit a Middle Bronze Age date, while the arrowheads belong to the Early Bronze Age, and the broken flint blade  was knapped in the Late Upper Palaeolithic. The extent to which these were heirlooms or chance finds kept or collected as curiosities is impossible to say, but there is no doubt that this combination of objects from different periods invokes, in some way, notions of ‘the past’. Could it be a foundation deposit associated with the overlying farmstead? If so, it was an act that symbolically laid the past to rest, and opened a new chapter at Horton.

This article appeared in issue 292 of Current Archaeology.


Leave a Reply

Your email address will not be published.