British Archaeology banner

Cover of British Archaeology

Issue 78

September 2004



Third Neolithic longhouse found in Scotland

Rare Medieval track excavated

Decorated shears trimmed Celtic hair

Iron Age 'bender' in Margate

Barrow saved from walkers

In Brief


Robert J Wallis and Jenny Blain report from the other side

Boscombe grave
The truth behind the latest Stonehenge Beaker finds

Digging up art
Clive Waddington reveals first dates for British rock art

Bronze bog hoard
Surprising objects in prehistoric hoard in Armagh

Seahenge story
Mark Brennand excavates on the beach

Forest fire
Peter Fowler describes new discoveries in the Languedoc


Ethnicity, mysticism, Roman disputes and hedges


Peter Drewett bemoans the lack of field skills


Neil Mortimer fights stone circle power on Ebay


Past Poetic: Archaeology in the Poetry of WB Yeats & Seamus Heaney by Christine Finn

Antiquaries: the Discovery of the Past in Eighteenth-Century Britain by Rosemary Sweet

Trethurgy: Excavations at Trethurgy Round, St Austell: Community & Status in Roman & Post-Roman Cornwall by Henrietta Quinnell

Urban Growth & the Medieval Church: Gloucester & Worcester by Nigel Baker & Richard Holt

Behaviour Behind Bones: the Zooarchaeology of Ritual, Religion, Status & Identity by Sharon Jones O’Day, Wim Van Neer & Anton Ervynck

Public Archaeology by Nick Merriman

The Victoria History of the Counties of England. Stafford vol IX: Burton-Upon-Trent by Institute of Historical Research

Archaeology, Ritual, Religion by Timothy Insoll

Charter Quay: the Archaeology of Kingston’s Riverside by Wessex Archaeology

Melrose Abbey by Richard Fawcett & Richard Oram

Places of Special Virtue: Megaliths in the Neolithic Landscape of Wales by Vicki Cummings & Alasdair Whittle

Human Evolution Cookbook by Harold L Dibble, Dan Williamson & Brad M Evans

CBA update

tv in ba

Looking back on a season of wars and battles


Chief archaeological scientist Sebastian Payne's new column

my archaeology

Philip Beale left his job for an archaeological experiment


ISSN 1357-4442

Editor Mike Pitts


This is why we dug Seahenge

The world knows about Seahenge. Archaeologists, however, have recorded not one, but two oak rings they call Holme I and Holme II. Mark Brennand, then with the Norfolk Archaeological Unit, directed the excavation at Holme-next-the-Sea. He tells his story

It is fitting that I should write here as the Holme timber circle project ends. The public saga began when Simon Denison’s British Archaeology report (News, December 1998) caught the eyes of Michael McCarthy, the Independent’s environment correspondent.

When McCarthy approached us, we were reluctant to talk. We thought the site was probably Middle Bronze Age (a palstave axe of that era had been found nearby), but we had no proof. To go public could have left us with egg on our faces. However, we helped McCarthy and crossed our fingers.

The story was on the Independent’s front page on 9 January 1999, a Saturday. The photograph of an ancient circle of posts surrounding an inverted tree stump caught people’s attention. There was no indication of scale: ‘Stonehenge of sea’ in the Independent’s title perhaps gave an exaggerated impression of size, and the rapid coining of ‘Seahenge’ introduced misleading associations.

Interest was international. Telephones at the Norfolk unit jammed; normal business was seriously disrupted. On the first weekend with a suitable low tide, police had to introduce a one-way system into the small village of Holme-next-the-Sea. The warden of the Holme wildlife reserve was overwhelmed with thousands of visitors.

Our excavation was a rescue project in every sense of the phrase. Erosion of this extremely fragile coastal landscape continues. Four years on, it is clear there would have been considerably less for us to study had we not intervened when we did.

The purpose of the excavation was not simply to remove the timbers (we could have done that in days), but also to retrieve as much information as possible regarding the sequence of construction and the contemporary environment. We knew from our evaluation that there were no visible prehistoric ground surfaces surviving inside the circle, and the exposed silt was considerably older than the timber structure. Our best hope of recovering informative material was from around the timbers themselves.

Posts had not been driven into the ground, but placed in a construction trench. Detecting its extent was a challenge indeed. A trial excavation in autumn 1998 had shown how difficult it was to see any stratigraphy. The excavation in plan of a putative construction feature was not going to be possible.

Instead, we cut a series of radially aligned sections, running from the inside of the circle, through the wall of posts and out to the area immediately beyond. Sometimes there seemed to be discolouration indicating the post to be set in the centre of a trench; elsewhere there was no visible evidence for construction at all. We took many samples from each section. The most compelling construction evidence consisted of dumps of timber debris from post splitting, thrown into the trench as it was being back-filled.

How was it made?

The excavations were challenging. Conditions were difficult and hours unsociable. The circle was only exposed at low tide, for two to five hours. Getting the trenches back to the state they were in on the previous day could take up half the shift. On top of this was pressure from the media, many visitors and a few individuals seeking to disrupt us. The excavations took twice as long as expected.

But curiously, for me the hardest part of the project was after the excavation: finding a way of interpreting the seemingly uninterpretable. I thought there would be a special deposit beneath the inverted tree that would explain the structure’s purpose. With hindsight this was unlikely, but what else was there to say? It felt like people were expecting great things that we were not going to be able to produce.

Post-excavation analysis has proved how wrong I was: there was a huge amount to glean about the circle and its surroundings. Maisie Taylor’s painstaking analysis of the timbers tells us about the tools, and thus about how the circle may have been constructed. Environmental work directed by Peter Murphy recreates the landscape setting and the later history of the site.

Construction began in the spring or early summer of 2049 bc, in the woodland where the trees were felled: at the circle site there is no debris from trimming branches and the ends of posts. Dendrochronologist Cathy Groves suggests there were 15 to 20 trees (but maybe as many as 26), all from the same location, which pollen indicates may have been quite close to the final site. Maisie believes that one tree was lying flat on the ground, blown by the wind or rocked over with the aid of ropes, before it was cut up into the inverted tree stump and four circle timbers.

It is here that the study of axe marks in the wood comes in. At least 51 different bronze axe blades have been detected undertaking felling, end-squaring and trimming branches; the actual number may be as high as 59. This does not take account of identical blades cast from the same mould, and we have no tool mark information from 11 of the 54 timbers. We believe that each axe represents a person. Indeed, it may be that many more people were involved who are archaeologically invisible.

Who built it?

Who were these builders, and what does this tell us about metal axe ownership in the Early Bronze Age? It seems clear that there was not a small group of individuals (specialist circle builders?) with a very large cache of various sized axes. If this were the case we would expect to see only the larger axes used for felling trees. This was the most common practice, but the largest, used on the central tree, did not fell any other trees. On more than one tree a larger axe was used for trimming side branches and ends than was used to fell the tree itself. Typically trees were felled by two axes wielded by two people working opposite each other, and then side branches and ends were trimmed by others, often with smaller axes. The use of different tools in different locations suggests that people were chosen for particular tasks.

The builders may have come from the local community, living in the immediate area of the monument, with axe possession relatively widespread. This may be supported by the potential division of labour seen at the felling site, perhaps dictated by gender with everyone present. This assumes that axe size was in some way related to the status or gender of the owner, and not general access to metal or axes.

The alternative hypothesis is that the builders were social elites from large families or dispersed communities. Everyone involved in the building may have travelled far to represent their group or family at the construction. Perhaps axes in Early Bronze Age Norfolk were possessed only by the most powerful or socially important individuals.

Complete tree trunks were moved to the site, which was almost certainly a saltmarsh in 2049 bc: insects from the construction trench suggest it was close to the high tide line, but not actually in the sea.

What did it look like?

Holes cut through the central stump and wear on one side suggest that it was pushed, pulled, hauled and dragged, with ropes manufactured from honeysuckle stems. The stump was then inverted into a pit, with the roots pointing upwards. This was aided by ropes that were trapped as the stump slid into the hole, not retrievable until it was removed over 4,000 years later. We presume that the construction trench was then excavated around the stump and the palisade of posts erected. The timbers’ excellent preservation suggests they went into waterlogged ground from the outset, so excavation of the trench may have necessitated bailing water. Debris suggests that timbers were piled up to the north-west of the circle, and manoeuvred into place after they had been split in half. Our interpretations of how this happened are based largely on the treatment and position of the posts themselves, which would of course not have survived under normal dry-land conditions.

Post 35=37 appears as two timbers above ground, but these are branches attached to a single trunk. The projected angle from the base of the fork suggests a gap would have remained, allowing a slim person to pass through. This is the only space in an otherwise solid wall of timber, and we suggest it is an entrance, even if only used briefly during the final stages of construction. Post 36 was placed outside the forked branch, blocking the view into the interior, but not necessarily preventing access. On the opposing side of the circle large split sections from the trunk of the inverted stump flank a single roundwood post (65).

Allowing for a low horizon to the north, the alignment from the gap between the branches of 35=37 to post 65 is remarkably close to the line of the rising midsummer sun to the north-east, and the setting midwinter sun to the south-west. We cannot know if this was intentional. Nonetheless, the south-western orientation of an ‘entrance’, the possible heightening of the timbers (they were set less deep in the ground than others), and the pattern of split and round wood all suggest that posts 35=37 and 65 were of prime importance. All the trees were felled in the spring or early summer, and it may be that the midsummer alignment was observed during construction.

We propose that the circle was then laid out in four panels. The forked timber and the single roundwood post centred two opposing panels or façades, to the north-east (all timbers from the trunk of the central stump) and south-west (between roundwood timber 40 and timber 30). Post 30 is the only one to display a split face on the outside of the circle. This reversal may be fortuitous, but we suggest that it was deliberate and deeply symbolic. The circle surrounds an inverted tree. The posts, with their bark to the outside, are mimicking a tree, except at this one point. The symbolism of this particular timber may have something to do with an individual or group involved in the construction, a deliberate break in the texture within the interior or exterior of the circle. The two façades were then joined by arcs of split timbers, to the north-west and south-east.

The timbers were set up to 1m in the ground: either they stood above ground for a greater height, or they were shorter and intended to bear the weight of an inner mound.

Peat around the outer edge of the post circle, which had begun to form within a century of its construction, suggests that no material was piled outside the timbers or spread from inside. Unlike the construction trench, that was almost immediately back-filled, a mound quarry ditch is likely to have remained open, with peat or organic material forming in it: but we found none of this. Thus we think the posts were at least 2 m high and the interior was not filled with earth.

If 51 people helped raise the circle, and maybe double that, it would have been an extremely busy scene, with a throng gathered round the stump, excavating the post trench and wrestling with half-split timbers. The site was crowded by a team of eight excavators and numerous specialists. The design we think we can see in the circle’s laying out suggests to us an ‘architect’ or overseer, but perhaps everyone was familiar with the principles and symbolism in the circle, even if they had never seen a structure quite like this before.

We could detect no further episodes of activity. Indeed, the structure’s completion may have been the only time all builders were on site together. It may be that they then simply walked away, their obligations fulfilled, and the feat became history. The communal building and the special treatment of the tree may have been the entire point. Elsewhere the construction of timber monuments may be but a single phase in a long and complex history, with timbers later being replaced in stone or covered with a mound. The Holme circle did not decay entirely, and was not afforded this treatment. Within 200 years peat began to build up around it, probably aiding preservation. Many centuries later, in the Middle and Late Bronze Age, the site was obviously still visible, and the focus for small gatherings that resulted in the accumulation of heat-shattered stones, pottery and animal bone, typical of burnt mound sites.

Why was it built?

The analysis has told us how the circle may have been erected, but not why. Early in the project, when asked by journalists, I simply spieled off theories of excarnation and an inverted underworld. I now feel less inclined to suggest a body was ever laid within the tree roots. Inversion, however, is literally central to the site. Elsewhere in the Early Bronze Age inversion (most commonly in funerary vessels) appears to be appropriate in some circumstances, but not in others. A pattern is not readily evident, but the symbolism may have transformed the everyday into the extraordinary. The tree stump was hauled from the place of its growth to a funerary monument in a saltmarsh, and turned upside down. It was then entirely closed off to the world around, possibly not even visible. There may be references to builders and astronomical events, and the large number of people involved emphasises the communal statement: but the main motivation seems to have been to place the tree stump in the right place, in the right way, within its own wall of oak posts.

While there are parts of the central stump’s trunk in the north-east section of the circle, most of that tree is not represented. Nor do we have all of the split posts, as there are too many half splits with the central pith intact. We cannot rule out that the missing timber was used in another structure. Where that might have been we can only guess, but we think the circle’s site was chosen for its proximity to an earlier monument, just 100 metres to the east.

This structure is represented by two concentric rings of roughly split oak timbers surrounding an oak hurdle-lined pit containing two large oak logs. There are good reasons for suggesting that the pit was a grave, and that the wood was all that remained of a palisaded barrow, not unlike examples excavated in the Netherlands. Radiocarbon dating indicates this monument was probably constructed several centuries before the excavated circle (c 2400-2030 bc), but it is possible that both monuments at some time were in use together.

Regular archaeological monitoring of the beach continues, and all visible structures have been recorded. Readers are reminded that the beach at Holme is a Site of Special Scientific Interest (SSSI) and a wetland of international importance under the Ramsar Convention, and private property. Now is breeding time for the large colonies of ground-nesting wading birds who are the beach’s summer residents; excessive visitor pressure would damage the wildlife and ecosystem.

Mark Brennand is regional research project coordinator, environment division, Cumbria County Council

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