British Archaeology banner

Cover of British Archaeology

Issue 70

May 2003



London graves descrated by Boudicca's army

Jacobite 'gentleman's retreat' at Glenfinnan

Bronze Age trackway and hide in East London

Evidence reveals peaceful Roman occupation of Scotland

How natives made money out of old Roman coins

In Brief


First humans in Britain
Nick Aston on the occupation of Britain during the Ice Age

Great sites
David Field on Europe's largest prehistoric mound, Silbury

Power dressing
Alison Sheridanon the significance of Bronze Age jewellery


Roman burials, Saxon immigration and an elephant cap


George Lambrick on the need for a National Heritage Act

Peter Ellis

Regular column


Mining in World History by Martin Lynch

The Past in Prehitoric Societies by Richard Bradley

Shaping Medieval landscapes by Tom Williamson

Novgorod edited by Mark Brisbane and David Gaimster

Raising the Dead by A J Stirland

The Geratest Killer by Donald R Hopkins

CBA update

favourite finds

Ros Niblet on an iron folding chair from Verulanium


ISSN 1357-4442

Editor Simon Denison


Great sites: Silbury Hill

Archaeologists have been intrigued by Europe's largest prehistoric mound for centuries. David Field reports on new work that overturns some old ideas

Great sites rarely come much greater than Silbury Hill - at least in terms of scale. This giant Neolithic tumulus near Avebury, the largest man-made prehistoric mound in Europe, has been a source of observation, speculation and wishful thinking for hundreds if not thousands of years.

Antiquaries and archaeologists have been visiting the site since at least the 17th century, and since then Silbury Hill has been interpreted variously as the burial place of an Ancient-British or Roman king, a platform for druid sacrifice, an astronomical observatory, a part of the Avebury temple, a temple to Mercury, a mound for assemblies and law making and a medieval motte.

Oddly enough, the first excavators of the mound, the Duke of Northumberland and Colonel Drax, were closer to the mark in 1776 when, announcing their intention of sinking a shaft 8ft square from the summit to the old ground surface in search of the contents, they issued what amounts to a press release. This clearly indicated that they believed the mound to be some 3,000-4,000 years old - with hindsight a not unreasonable estimate.

Again and again excavators returned, hoping to solve Silbury's mysteries - Blandford and Faulkner in 1849, the Rev Wilkinson in 1867, AC Pass in 1886, and Flinders Petrie just after World War I. But until recently the standard interpretation was that of Richard Atkinson, formed as a result of his televised excavations from 1968 to 1970, made with the bbc.

Atkinson dug trenches extensively - on the summit, down the slopes and across the ditch. He reopened the tunnel, first dug in 1849, along the old ground surface to the centre of the mound. If he, or his television sponsors, were hoping for gold, or even for a single decent burial with human bones, they were disappointed.

But he did reach conclusions about the structure of the hill, which he believed to comprise an initial smallish mound of gravel and turf, surmounted later by a second mound of chalk with a surrounding ditch, over which was constructed the massive chalk mound that we see today. The final mound, he thought, was built in tiers like a wedding cake, arguing that the now-invisible lower terraces had become obscured by weathering. Material from the tunnel suggested a Neolithic date - but none of his radiocarbon dates were safe by today's standards.

The sudden collapse of the 1776 shaft at the summit in 2000 prompted a new phase of research on the hill - including various forms of non-invasive survey, seismic survey and limited excavation. Some of the initial results of this English Heritage work have received press publicity over the past two years, but the analysis has continued, shedding light on many of Atkinson's conclusions and the history of the hill.

Cores bored for the seismic survey indicated that no major turf line developed between Atkinson's construction phases, refuting the theory that each phase of the mound was left to stand for years before being remodelled. Analysis of the cores also found that turf on the old ground surface had been trampled and mixed with chalk - an evocative insight into the busy activity involved in the mound's original construction.

Meanwhile, two small trenches cut on the summit by Fachtna McAvoy of English Heritage's Centre for Archaeology revealed - for the first time -a fragment of antler from a secure context, lying against a chalk wall in a deposit of chalk rubble. This produced a secure radiocarbon date of between 2490-2340 BC, placing the mound firmly in the Late Neolithic.

Outline surveys of the monument have been made in the past, and Atkinson produced a photogrammetric contour plan from air photographs - but, perhaps surprisingly, no full survey of the earthworks had been undertaken before the English Heritage work of the past three years. The breaks of slope, the ledges, platforms and scars that are visible to the naked eye had not been recorded. As a result of the new work, we now know that the mound is not circular, but that each level is a polyhedron with a number of straight sides - as many as nine at the base, possibly fewer towards the summit.

The mound that we see today is the product of several thousand years of adaptation, use, weathering and erosion. Like any parish church, its earliest features are partially obscured by those of more recent date.

Many of the scars on the summit and slopes are the result of past excavations. The spoilheap around the 1776 shaft - still prominently visible in the mid-19th century - has almost vanished, but the location of Atkinson's trenches can be seen quite clearly. Crowded fairs took place on the hill in the 18th century and they might account for some of the dips and banks. But the winding path to the summit appears to be in the same position as that noted by John Aubrey in his sketch made around 1665, so any scarps that the path cuts must predate it.

What of Atkinson's contention that the hill was raised in tiers? The new survey work suggests otherwise. The uppermost terrace, when circumnavigated, returns to a point several metres below the starting point - in other words, it spirals down. Weathering has obscured the detail of ledges further down the slope; but if we assume the feature is continuous, it implies a spiral path all the way from the summit to the base.

The idea of a spiral path is attractive. It would not only have provided a processional route to the top of the hill, but would also have aided the construction of the mound. Atkinson's tiered structure poses enormous logistical problems for anyone hauling material to the higher levels. Not impossible of course; and great amounts of scaffolding, ladders, gantries and other construction paraphernalia could have been used to help. However, a spiral would have allowed easier access, and material could have been carried or dragged up to the working level with considerably less fuss.

So what evidence is there of a Neolithic date for this spiral path? In fact very little. Atkinson did find a portion of chalk rubble walling on the inside edge of the upper terrace, which he claimed was similar to walling on the summit; and as it appeared to extend into the mound, he clung to a view that the terraces were part of the initial construction. Some terraces visible today are clearly post-Neolithic, as they cut into the mound's original profile. But we have no compelling evidence to reject the idea that some type of spiral path or series of terraces did indeed form part of the original design.

One intriguing aspect of Atkinson's and the earlier excavations at Silbury is that many more Roman and early medieval finds were produced than Neolithic. In the 19th century, a substantial Roman building was found south of the hill. Cutting into the mound at ground level in 1867, Wilkinson found a platform just below the surface on which was a pile of ashes associated with Romano-British artefacts. Atkinson himself found over 100 Roman coins in the ditch, while numerous Roman shafts and wells were found nearby. From this evidence alone, it was not surprising that some researchers have claimed that Silbury was a sacred site in the Roman period.

Then, over the past decade, the Roman evidence increased. A combination of pipeline excavations and aerial photography has revealed the extent of Romano-British settlement on the lower slopes of Waden Hill - a natural hill next to Silbury. It now seems almost inconceivable that, with a Romano-British settlement facing the mound, Silbury was not used in some way. It may be that the mound was once covered with burials, monuments and memorials, and that what has been found so far is the tip of the iceberg.

Atkinson's medieval evidence was no less striking. Small postholes containing iron nails, early medieval potsherds and a silver coin of Ethelred II dating to 1010 suggested that the terraces had been revetted by posts. He also found an iron spearhead, and explained the evidence as defence against the Danish invasions - not unreasonably, given the skirmish at nearby East Kennet in 1006 reported in the Anglo-Saxon Chronicle. He concluded that the original terraces had been recut and fortified. The postholes, however, are located on the inside of each terrace and therefore imply revetment rather than defence.

It seems more likely, perhaps, that Silbury Hill was used as the site for a prestigious building, perhaps a Christian building, in the early medieval period. If the mound had been sacred in the Roman era, it is possible that it retained its religious attraction in later centuries.

Locally, the struggle to establish Christianity appears to have been quite vigorous. The presence of a Saxon cross, followed by the founding of a Christian church at Avebury - within sight of Silbury - emphasises the enormous interest in the area during the conversion period. Avebury subsequently attracted a monastic cell, and suffered attempts to destroy its standing stones. Given this concern, and indeed papal encouragement, to Christianise pagan monuments, it would be no surprise if Silbury Hill also saw some form of Christian modification. Sadly no direct evidence of Christianity has yet been found.

The ditch surrounding Silbury Hill is often considered a mere quarry from which the mound material was derived. However, its circular nature, and the regularity of its rectangular western extension, indicate that it served more than a functional purpose.

Archaeologists have come to see that ditches, even massive ditches around henges or hillforts, need not always be just utilitarian structures but may have had a metaphysical function too - for example, to keep evil spirits at bay. The rectangular extension at Silbury, if waterfilled, would have served as a cistern or reservoir. Elsewhere in the world, cisterns have often been the focus of ritual and ceremony. The mirror-like quality of standing water may have had symbolic implications too. Given archaeologists' fascination with shamanism, it is significant that mirrors are considered symbols of shamanic ceremony and power.

For just three days in early summer 2001, as the water-filled ditch dried out, a huge vegetation mark, straight-edged and some 10m wide, appeared to extend across the ditch floor for some 50m towards the mound. Its orientation, however, was curious, running diagonally across the ditch extension towards a position off-centre of the mound. The feature definitely seems man-made. It may be that the hill's Neolithic builders dug a deeper channel here to collect water from local springs and bring it to the deep ditch encircling the mound.

It seems likely, then, that Silbury Hill's ditches were intentionally filled with water. Furthermore, the hill itself was built next to water, close to the River Kennet. The siting of this monumental mound in a valley - so that its summit barely attains the level of the surrounding hilltops - has often raised comment. Why not build higher up? The answer must be that the place itself was as important as the mound.

Archaeologists have increasingly noted how many Neolithic and Bronze Age monuments are built in 'liminal' or cross-over points in the landscape, and close to water (ba, April 1999). The well-known monuments at Newgrange in Ireland, for example, cluster alongside the River Boyne. There are numerous other examples. Silbury is not only close to the River Kennet, but also at a geologically liminal position, on the very edge of the chalk where it abuts the river's flood plain.

Why was such a position considered important? And why build this unparalleled, monumental mound? We will never have certain answers to these questions. But the proximity of earlier monuments such as West Kennet long barrow and the Windmill Hill causewayed enclosure shows that Neolithic people had built monuments in the area for generations.

The place had an ancestral draw. The significance is obscure, and we can only hope to understand it if we combine our scientific approach with a vision that sees beyond modern-day beliefs about how landscape must have been viewed and used in prehistory.

David Field is a field archaeologist with English Heritage, and was responsible for the recent landscape survey of Silbury Hill

CBA web:

British Archaeology

Jan/Feb 2005
Mar/Apr 2005
May/Jun 2005
Jul/Aug 2005
Sep/Oct 2005
Nov/Dec 2005
Jan/Feb 2006
Mar/Apr 2006
May/Jun 2006
Jul/Aug 2006
Sep/Oct 2006
Nov/Dec 2006
Jan/Feb 2007
Mar/Apr 2007
May/Jun 2007
Jul/Aug 2007
Sep/Oct 2007
Nov/Dec 2007
Jan/Feb 2008
Mar/Apr 2008
May/Jun 2008
Jul/Aug 2008
Sep/Oct 2008
Nov/Dec 2008
Jan/Feb 2009
Mar/Apr 2009
May/Jun 2009
Jul/Aug 2009
Sep/Oct 2009
Nov/Dec 2009
Jan/Feb 2010
Mar/Apr 2010
May/Jun 2010
Jul/Aug 2010
Sep/Oct 2010
Nov/Dec 2010
Jan/Feb 2011
Mar/Apr 2011
May/Jun 2011
Jul/Aug 2011
Sep/Oct 2011
Nov/Dec 2011
Jan/Feb 2012
Mar/Apr 2012

CBA Briefing

Courses & lectures
CBA Network
Grants & awards

CBA homepage