Unravelling the evolution of Europe’s largest prehistoric mound
When a tunnel into Silbury Hill was opened for the first time in almost 40 years to allow emergency conservation work, a team of English Heritage archaeologists seized the opportunity to enter the mound. Their recently published work has revolutionised our view of this magnificent prehistoric monument, as Jim Leary told Matthew Symonds.
Silbury Hill, seen from the south-east. Recent stabilisation work inside the mound allowed a team of archaeologists to gather new information about how the monument was constructed.
On the Late May Bank Holiday weekend in 2000, a visitor surreptitiously slipped over the fence at the base of Silbury Hill and set about scaling its flanks. Although public access to the summit was officially banned in the 1970s, many still made the ascent, illicitly taking the opportunity to conquer Europe’s largest prehistoric mound. From the top, they could look out over the West Kennet valley and its famous prehistoric monuments. On this occasion, though, it was not the view that captivated the visitor so much as an ominous hole that had opened near the summit’s centre. This disturbing discovery would, ultimately, lead to the demolition of many modern myths about Silbury Hill.
Fixing a hole
The visitor contacted the National Trust, who in turn alerted English Heritage. A temporary corrugated-steel shelter was hastily erected over the chasm, and a video camera lowered into its depths. Once the recording revealed the smooth edges of a vertical shaft descending into the mound, the reason for the collapse became clear. In 1776, Colonel Edward Drax had employed a group of miners to burrow into the heart of Silbury Hill. Although few details about this episode were ever published, beyond Drax finding no more than a ‘thin slip of oakwood’ at its base, it was always assumed the shaft had been adequately backfilled. Now its dramatic reopening suggested otherwise. As the void gradually expanded over the winter months, with its mouth measuring 40m ² by March 2001, the search for an explanation gave way to a far more delicate question. What should be done?
An aerial photograph of Silbury Hill on 30 May 2000, shortly after the hole was discovered, and the day before it was covered with scaffolding. The void is visible near the centre of the summit. It soon became apparent that the collapse was a consequence of Drax’s 1776 diggings opening up again, which explained the shaft’s smooth sides.
To buy time, the hole was packed with lightweight polystyrene blocks, while English Heritage weighed up the implications of previous investigation into Silbury Hill. The Drax shaft was only one of a series of cavities cut into the monument over the preceding two and a half centuries. In 1849, the Very Reverend John Merewether drove a horizontal tunnel into the centre of the mound, a feat repeated on a far more ambitious scale between 1968 and 1970 by Richard Atkinson, and memorably broadcast live by the BBC. The slumping of the Drax shaft capping raised the alarming question of whether the other tunnels riddling the mound had been properly backfilled. To find out, bore-holes were sunk over the 1968-1970 tunnel, and a camera lowered in. This swiftly established that there were still gaping voids along the course of the tunnel, forcing English Heritage to confront the nightmare scenario that the mound itself could collapse.
The search for solutions was a long one, but ultimately only one course of action was open. ‘When the Atkinson excavations ended in 1970,’ Jim Leary, English Heritage’s director of fieldwork at Silbury Hill from 2007 to 2008, explains, ‘they simply closed the tunnel door and piled soil against it. The various engineers consulted agreed that the only way to resolve the problem was to open that door again, re-enter the tunnel, and backfill it properly.’ This conservation headache offered the prospect of an archaeological silver lining, though. The physical scars of excavation work within Silbury Hill were far more impressive than their instigators’ literary legacy. When Alasdair Whittle wrote up Atkinson’s unpublished records, he was faced with the task of piecing together a very fragmented archive. Now that the tunnel had to be reopened, it provided an opportunity to harvest fresh data and test Atkinson’s conclusions. ‘No one would ever have had permission to go back in and re-examine the mound from the inside otherwise,’ says Jim. ‘The series of events were awful, dramatic, and caused damage to the monument, but ultimately something positive has come out of them.’
The 1968 contour plan of Silbury Hill, showing the course of the 1968-70 tunnels, and location of the surface trenches.
Atkinson saw Silbury Hill as the product of four broad phases of construction. With such a small number of steps between conception and completion, he believed the monument could be explained as the grandiose vanity project of an individual chieftain or religious leader who had a clear idea from the outset of the indelible statement he or she wanted to inscribe on the landscape. The English Heritage team who entered the mound in 2007 were restricted to gleaning what they could from the stratigraphy exposed in the tunnel sides and a very limited amount of fresh excavation. To their surprise, the results suggested a very different origin for Silbury Hill.
Over the winter of 2000/2001 the area of collapse widened. Here an English Heritage archaeologist abseils into the void to collect a sample.
‘We expected to find Atkinson’s four phases of activity,’ Jim remembers, ‘because that was the received wisdom. But instead we found clear evidence that the mound evolved gradually, year after year, almost like a continuous project. I do not think that a year would have passed without some kind of work or activity on the hill. But we shouldn’t see this as a concerted attempt to create a massive, imposing monument, because the early phases were very, very small. One of them was a knee-high gravel mound. So if someone was trying to make a statement, you have to wonder what it could be. I think we need to look beyond this notion that Silbury Hill was all about showing off power.’
The earliest phases of the mound — if indeed ‘phase’ is the right way to describe a process that appears to have been essentially continuous — were certainly fairly unprepossessing. As the monument rests directly on clay soil, the turf must have been carefully stripped away before construction began. The bare soil was then churned up by large numbers of feet, and perhaps hooves, suggesting that the site was the venue for some kind of intensive activity before work started on the earliest iteration of Silbury Hill.
Rather than laying a foundation stone, the first act was to pile gravel into a modest mound. As this gravel heap grew in size, it was ringed by a circle of wooden stakes, while further piles of soil lay beyond like orbiting satellites. This cluster of features was then capped by a mass of earth and turf. A step-change in the monument’s ambition seems to occur when a bank and ditch about 100m in diameter was constructed around the mound. This was enlarged with chalk hardcore, and — as more chalk, clay, and other materials were gradually heaped on top — the mound rose higher and higher above the West Kennet valley.
The entrance to Atkinson’s tunnel into the heart of Silbury Hill, seen in October 2006. The door was still padlocked shut, but the key was found lying underneath! Many voids were left inside after the 1960s campaign, as this view of the main tunnel shows.
It is not only the mound’s convoluted construction that challenges the traditional model of a powerful leader pursuing a developed master-plan, but also the refined dating evidence the English Heritage team secured. This reveals that work on Silbury Hill began around 2,450 BC, and continued for between 55 and 155 years. ‘This has important implications for how we interpret the monument,’ says Jim. ‘If you take the midpoint, which is a little over a century, we are looking at about four or five generations of people to go from nothing, to the mound itself. So this does not fit notions of a single figure acting as a driving force. Equally, raising the mound didn’t take centuries and centuries. As dating techniques improve, we’re finding that these Late Neolithic monuments have lives that span generations. Nearby, Stonehenge provides an especially enduring example, and its great sarsen trilithons were being raised at around the time Silbury Hill was built.’
A schematic diagram showing the various phases in the construction of the Silbury Hill mound, and the course of the 1776 shaft and 1960s tunnel. Rather than progressing in an orderly sequence of four phases, as Atkinson argued, the new study reveals that the mound gradually evolved through a process involving multiple phases, not all of which aimed at creating a mound.
The length of time lavished on manufacturing Silbury Hill also begs the question of whether the people who laid that first pile of gravel on the site were setting out to build a monument similar to the one visible today, or if they were pursuing a different agenda. ‘I don’t think it was started with that final form in mind,’ Jim says. ‘For that to happen, the last generation working on the mound would have to know what their great, or great-great, grandparents were visualising when they started it. We can see that the monument itself morphed and changed over the course of decades. It’s important to remember that not every phase is a mound — at one point a banked enclosure is the main feature. I think we have become too hung-up on the final form of the monument, and forgotten about the process of creating it. Just look at Medieval cathedrals: they were never constructed with the final form in mind. They were built in a piecemeal fashion, with every generation leaving their own mark, some of which look quite odd.’
Divorcing the act of building from the monument’s current appearance is a radical step. Previously, attempts have been made to explain the rationale behind Silbury Hill by focusing on where it was visible within the landscape and how it impacted on the skyline. If the English Heritage team are right, then such attributes are no more than a coincidental by-product of how the mound evolved. This conclusion is reinforced by the discovery that the mound’s iconic flat-topped profile is probably a consequence of more recent landscaping. The original Neolithic summit appears to have been truncated in the Saxon period, when an 11th-century fortification was erected on the hill.
Bringing the landscape together
Jim believes that obsessing about the hill’s form has blinded us to a far more important element: the material making up the mound. ‘I think people were deliberately bringing parts of the landscape together,’ he says. ‘It could almost be seen as a literal, physical composition of their place — a monument to the landscape they live in.’ The variety of material incorporated within the mound is certainly intriguing. As well as turfs cut from the clay soil on which the mound stands, quite possibly the very ones stripped off before construction started, grass that had grown on chalky soil was also brought to the site from further afield.
Estimates of the construction dates for various elements of Silbury Hill, based on the radiocarbon dates secured from the mound.
Sarsen boulders have also been found in the mound. Such stones are scarce in the immediate vicinity of Silbury today, but it was likely that they were far more common in the prehistoric period. Yet even though the stones may not have travelled far, they should not be dismissed as merely a convenient source of hardcore. Some boulders found high up in the mound would have needed a couple of people to manhandle them up the incline, making them far from the most convenient building material. Among this potent mix of local produce, space seems to have been made for at least one example of a rather more exotic stone type. A fragment of rock from the mongrel ‘bluestone’ geology that famously made its way from the Welsh Preseli hills to Stonehenge was found on the summit by Atkinson. Sadly, the bluestone lay loose in the topsoil, leaving its role a mystery. To this can now be added three further small ‘bluestone’ flakes from the 2007 summit excavations, but none of these were within secure Neolithic contexts either. One possibility is that a monolith once crowned the mound; another is that its presence is a product of the destruction of some Stonehenge bluestones during the Neolithic period.
A reconstruction drawing of construction under way on the upper mound of Silbury Hill in the Neolithic period.
The level of preservation exhibited by the Neolithic turfs sealed within Silbury Hill is one of the most remarkable features of the monument. ‘It is absolutely phenomenal,’ Jim says. ‘In some instances the grass is still green, and there is moss that we can say grew on a north-facing slope. There are also a lot of astonishingly wellpreserved insects, so we are getting wonderful snapshots of the environment at this time.’ Atkinson’s discovery of winged ants led him to propose that Silbury Hill was built at the beginning of August, when such insects take to the air. Jim, though, is not convinced. ‘It would be wonderful if that was the case, but while Atkinson found ants, and wings, he didn’t find them on the ants. We just don’t know what time of year people were working on the mound, so I think it’s time to move on from any idea of a “harvest hill”.’
Plant material from within Silbury Hill; in some instances, it still survives as a thin strand of green. Such remarkable preservation has given exceptional insight into the Neolithic environment.
Even if the final form of Silbury Hill is a red herring, the monument’s location is likely to be significant. ‘The area had been a focus for activity since the Early Neolithic,’ Jim points out. ‘It is overlooked by the Windmill Hill causewayed enclosure on one side and the West Kennet longbarrow on the other, so it is effectively an ancestral landscape. But I think it would have been important anyway, because it is right at the head of the Kennet River, which flows directly into the Thames. Today, we say the Thames flows up through Oxford, but that’s a result of modern cartography making the longest tributary the head of the river. I think Neolithic people could easily have seen the River Kennet and the River Thames as one and the same. The largest henge in the country, at Marden, lies at the head of the River Avon, so — if Silbury was considered the head of the Thames — we are looking at a focus on the sources of major sacred rivers.’
If Jim is right, and the communal act of building was more important than the final form of the monument, was it ever completed? ‘I think rather than being finished, people just stopped working on it,’ says Jim. ‘There was no end, as such — the world just moved on and no one added to it any more. This period in time coincides with when the first Beaker pottery, Beakerstyle burials, and copper arrive. So it is a period of very great change. Silbury could be a reaction to that by the old guard, but ultimately the Beaker people won the day, and came to dominate the archaeological record.’
Apart from a few fragments of later Beaker pottery, dating to centuries after work on the mound ceased, there is very little sign of later prehistoric activity at the site. It was not until the Roman period that there was another burst of activity at the source of the Kennet, when a small town grew up in the shadow of Silbury Hill. There is surprisingly little sign of Roman disturbance to the mound itself. Did the townsfolk sheltering by the monument’s flanks revere and avoid it, or could it be that a Roman shrine on its summit was destroyed in the next flurry of activity? During the 11th century, a spiral walkway was carved into the mound. Littered with Saxon pottery and a coin of Ã†thelred the Unready, this adaptation can probably be paired with the postholes on the summit. These point to an early 11th-century fortification having been constructed on the hilltop. The Anglo-Saxon Chronicle records a surge in Viking raiding in the area at that time, and it seems likely that the local population found a more practical purpose for the mysterious mound they had inherited.
After the Saxon episode, the most obvious scars left on the hill by later generations were created by antiquarians and archaeologists. Looked at that way, it can be questioned whether work on the monument ever truly ended. ‘Publication of our results provides this generation’s statement about Silbury Hill,’ says Jim. ‘But every generation creates the hill that they want. In that sense, it is still serving its purpose to this day.’
A reconstruction drawing of Silbury Hill in the Medieval period. This shows an 11th-century fortification perched on the freshly truncated summit, suggesting that the local population had recast the mound as a refuge from Viking raids.
This article appeared in issue 293 of Current Archaeology.