The voice of archaeology in Britain and beyond

Cover of British Archaeology 92

Issue 92

January/February 2007



Ancient trading power near Inverness

Rare insights into a medieval city

Possible new neolithic enclosure on Orkney

Roman Colchester unveils more of circus

In Brief & Phase 2


Stonehenge Douai manuscript discovered
Christian Heck describes his surprise Stonehenge find, a new medieval depiction

Transit van excavation
Bristol students find more than bunk in an old Ford van

Are these the pyramids to revolutionise Europe?
Anthony Harding investigates date claims of 12,000 BC/BP near Sarajevo, Bosnia.

Among tombs and stone circles on Banc Du
A history of discover of Neolithic earthworks and causewayed enclosures

on the web

Recommended websites


Views and responses

CBA news

Headlines from the CBA office


ISSN 1357-4442

Editor Mike Pitts


Among tombs and stone circles on Banc Du

The UK's first neolithic earthwork enclosure was recognised nearly a century ago. Since then many have been found: but only now has one been proven in Wales. Timothy Darvill, Geoffrey Wainwright and Toby Driver explain why their Pembrokeshire excavation may lead to new insights into some of the great questions about prehistoric Britain.

The existence of neolithic enclosures in Wales has long been suspected, but only recently confirmed. When Cecil Curwen first discussed Britain's neolithic camps in Antiquity for 1930 he included one possible example west of Offa's Dyke, Dinas near Llanidloes in Montgomeryshire, but subsequent work has shown it to be later in origin. With attention focused on the great causewayed enclosures of central southern Britain – such as Windmill Hill in Wiltshire and the Trundle in Sussex – very little interest was shown in searching western landscapes, despite the widespread occurrence of portal dolmens, passage graves and long barrows within and around the uplands.

Roger Mercer's investigations at Carn Brea in Cornwall in 1970–3 broke the mould not only by showing that a simple walled enclosure at the site was built in the fourth millennium BC, but also that not all neolithic enclosures were causewayed. Other similar sites were soon recognised in Cornwall, and two or three possible examples were noted in south-west Wales, for example Clegyr Boia, Garn Fawr and Carningli. At a meeting of the Neolithic Studies Group in London in 1998 something of the great range in shape, size and construction details of fourth millennium BC enclosures was clear. Aerial photography, fieldwalking and sample excavations in Wales continued to identify possible and likely neolithic enclosures, some showing similarities with causewayed enclosures and some not, but the associated material culture was either poor or loosely tied to the boundary works while absolute dating was missing.

During a flight across the central Preselis of north Pembrokeshire in July 1990 Chris Musson first spotted an unusual looking hilltop enclosure on Banc Du near the New Inn crossroads, Morvil. It was not until December 2002 that Toby Driver was able to photograph it again under suitable conditions for archaeological interpretation. Amid a series of cultivation features and abandoned trackways were two non-concentric and incomplete earthworks defining a roughly oval area around the hilltop. Site inspection confirmed that parts of the earthwork seemed to be causewayed, and in a relatively protected area on the north-east side of the hill the bank clearly comprised a series of discrete segments. With the generous permission of the Davies brothers of Brynberian, who own the site, topographic and geophysical surveys have been undertaken to map the archaeology, and an evaluation trench excavated across the inner earthwork.

Banc Du is a prominent, fairly flat-topped southwards-projecting promontory of Foel Eryr, at about 334m od at the west end of the Preseli ridge. There is a crag-line on the south-east side and steep slopes to the south and west sides. It overlooks the source of the Afon Syfynwy to the south-east, and affords extensive views to the south and west. Neither upland nor lowland, Banc Du occupies that curious middle position between enclosed fields and open moor, sometimes one sometimes the other, clearly cultivated in the past but now maintained as improved pasture for grazing sheep.

The surveys suggest three main period of activity across the hill. The most recent comprises extensive cultivation in a kind of ridge and furrow pattern that is probably associated with a now-abandoned, putatively medieval or later, long-house settlement or small hamlet downslope to the north-west. Earlier than this are two curved trackways crossing the hilltop that each seem to provide access to a single long-house dwelling or structure of some kind, although whether these are contemporary or successive is unknown.

Peeling away these features we are left with the earthworks of the earliest visible phase, the seemingly interrupted banks and ditches noted by Musson and Driver. It has not yet been possible to map the full extent of these using geophysical survey, as the underlying Ordovician shales are not especially responsive and medieval or later cultivation is masking some earlier features. However, by combining the results of all the surveys and remote sensing so far available, a provisional plan can be established. This shows an inner boundary defining a slightly oval space that is perhaps open against the cragline on the east side. Outside is an outer boundary that again seems to start on the cragline, crosses the neck of the promontory on the north side and then curves round on the west side to run close and parallel to the inner boundary along the hill's south side.

In August 2005 we excavated a single evaluation trench, measuring 15m by 1m, across the inner earthwork on the north side. The bank, about 3.8m wide, was preserved to a height of c40cm, and appears to have been set on an old land surface that had been deturfed. A posthole near the front edge of the bank and a pair towards the rear is suggestive of some kind of timber lacing within the bank. A single stone on a terrace in front of the bank together with a quantity of large, fresh stones at the bottom of the ditch suggest stone facing. The ditch was 2.8m wide and a maximum of 1m deep, and slightly U-shaped in profile. The main fills were sealed by a heavy iron-pan. The primary fill comprised stony deposits resulting perhaps from the early collapse of the stone facing above. Over this were two dark, ash-rich layers separated by clean clayey deposits. Plants represented by charcoal from these layers included heather, hazel, alder and oak. The upper fills comprised layers of yellow stony clay and grey silty clay from a variety of sources, including the bank.

No cultural material was found in the small section explored, but six radiocarbon dates were obtained through the ahrc-funded Neolithic Enclosure Dating Project coordinated by Alasdair Whittle and Frances Healy. These showed that the initial silt of the ditch accumulated at around 3650BC, while the middle fills overlying the stone collapse contain material from the period 3000–2600BC. Ralph Fyfe has examined pollen from the old ground surface under the bank, and concludes that the enclosure was built in a heath-dominated landscape with some local scrubby woodland including oak and hazel.

Although small scale, and still continuing, the investigations at Banc Du have profound implications for understanding occupation of the area in the fourth and third millennia BC. Not only is this the first confirmed neolithic enclosure in Wales and the mid west of Britain, but it is also notable for the fact that visible earthworks survive at ground level. This offers hope for fieldwork elsewhere in the region. It is an incentive to check other hilltops carefully and to examine closely other recorded but poorly studied earthwork enclosures that are so often arbitrarily assigned to later prehistory with no justification whatever.

At Banc Du itself there is clearly much to learn about the nature and use of the site. The presence of postholes within the bank and evidence for a stone facing on the outside suggest a fairly sophisticated structure. Rough timber lacing has been found in the banks of a handful of neolithic enclosures elsewhere in Britain, notably at the Stepleton enclosure within the Hambledon Hill complex in Dorset. Timber revetment is also a feature of many continental enclosures of the fourth millennium BC.

Banc Du's greatest significance is probably in relation to its contemporary landscape. When first built, the enclosure would have been contemporary with the great megalithic tombs of the area such as the portal dolmen later incorporated into a long barrow at Pentre Ifan just 7.5km to the north, and the passage grave at Bedd yr Afanc 6km to the north-east. This was also the period that fine metamorphic rocks began to be exploited in the area for the manufacture of axes: petrological Group VIII, for example, whose products have been found widely scattered across south Wales, and southern and midland England.

The later use of the enclosure in the second quarter of the third millennium BC is perhaps even more significant on a wider canvas. Banc Du lies towards the north-western end of the main Preseli ridge that provided spotted dolerite and other kinds of bluestone used in megalithic monuments. The outcrops at Carn Menyn, enclosed by a prehistoric ditch, are just 8km along the ridge from Banc Du (Beyond Stonehenge: Carn Menyn and the bluestones, Jul/Aug 2005).

Several stone circles in south-west Wales were constructed from these rocks. The ridge also provided the 80 or so pillars taken 250km eastwards to Stonehenge and built into the double circle forming phase 3I soon after 2600BC. These stones were reused in all subsequent remodellings of Stonehenge, and still cause interest and controversy to this day.

Clearly they were important to the people of third millennium BC Wessex. Understanding why that should be so depends on identifying the social context of the outcrops from which they originally came. That means following the bluestone trail back to the Preselis, and unpicking the landscape there. Banc Du now provides a secure foothold from which to explore those worlds.

Timothy Darvill is head of the Archaeology and Historic Environment Group, Bournemouth University; Geoffrey Wainwright is treasurer of the Society of Antiquaries of London and former English Heritage chief archaeologist; Toby Driver is project manager for the aerial survey programme at the Royal Commission on the Ancient and Historical Monuments of Wales. They would like thank Louise Barker, Paul Cheetham, Vanessa Constant, the Davies family, Roger Doonan, Ralph Fyfe, Frances Healy, Ehren Milner, Chris Musson, David Percival, Tom Pert, Bronwen Russell, Yvette Staelens, Anna Stocks, Judith Wainwright, Alasdair Whittle and Georgina Wilson for help with the project. For the 1998 Neolithic Studies Group meeting see T Darvill and J Thomas (eds) 2001 Neolithic Enclosures in Atlantic Northwest Europe (Oxbow)

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