The voice of archaeology in Britain and beyond

Cover of British Archaeology 107

Issue 107

July / August 2009



Scottish dig has big surprise in the post

Urine to navel fluff: the first complete witch bottle

Celtic tankard adds value to Welsh treasure

In the press

In Brief & Phase 2


on the web

Recommended websites
Websites of univeristy archaeology departments and a community site for Digging Vindolanda.


your views and responses

CBA Correspondent

Some recent projects benefiting from Challenge Funding.

my archaeology

Simon McBurney is a writer and actor; his father, archaeologist Charles


ISSN 1357-4442

Editor Mike Pitts


THE BIG DIG: Hambledon Hill

Earthwork enclosures with a seemingly unnecessary number of causeways across their ditches have long been attributed to Britain's earliest farming era, dating back to 4000BC. Yet their purpose remains obscure. Roger Mercer directed a classic excavation at one such site in Dorset over 20 years ago, now newly published. Here he describes the insights established by what he found.

By 3300BC, the sprawling hilltop above the confluence of the rivers Iwerne and Stour supported a complex of monumental earthworks that included two long barrows, two enclosures and multiple outworks, perhaps one of the largest of its kind in Britain. The scale of the long ditches and timber-faced walls seems partly attributable to the fear of attack. Within the four previous centuries, burning ramparts were brought down at least three times; we found the remains of people buried under the rubble, accompanied by the flint arrowheads that may have killed them. Yet for all its size, this complex was not a town, or even a village. Occupation was seasonal and light. Human skulls on the bottoms of ditches, and other carefully-placed deposits, indicate that domestic existence was far from the sole interest of the people who looked out across the Dorset landscape from the top of Hambledon Hill.

All this we know from the results of major excavations which began in 1974, following the realisation that ploughing was badly eroding what little seemed to remain of the earthworks. When excavation ended in 1986, many hundreds of mostly young archaeologists had been on site for a total of 102 weeks. We excavated numerous ditches and pits, greatly extended the known size of the complex and recovered vast quantities of broken pottery, flint tools, animal and human bones and other finds. In the analysis that followed, discoveries included a grape pip and a single piece of mature vine charcoal (implying vines were grown in the area), the earliest known evidence for Vitis vinifera in Britain and the oldest such charcoal in northern Europe. There were also a few fragments of Erica vagans, the only known instance from prehistoric Britain of this heather that today is found exclusively on the Lizard in Cornwall – whence came also some fine neolithic pots.

The full details of this research are set out in two new volumes (see end note), which describe a landscape and the development of its personality over ten millennia. I think I can speak for all the contributors in expressing relief and enormous pleasure at the report's public appearance. What I will do here is lay out a little of the extraordinary neolithic story.

What remains

The gently contoured Hambledon Hill stands between the chalk mass of Cranborne Chase to the east – clearly visible from its western slopes at only 3–4km distance – and the rest of the Dorset chalk to the south and southwest. On a good day, you can see the Mendips to the west and as far as the Isle of Wight to the southeast. Yet hidden by the larger chalk massifs, the hill's true shape and volume is apparent only from the flat land of the west and north. Hambledon is a "secretive landmark" so typical of the neolithic.

Aerial archaeology has revealed two successive field systems on the hill, and excavation has shown it was ploughed from at least 1800–1500BC to the end of the Roman period. Some of the most punishing farming occurred much later, between 1960 and 1980. On wet autumn days, roads near the hill are today covered in latte-like fluid, and the rivers become swollen and milkybrown in colour. Agriculture has greatly augmented the natural processes of weathering and solution.

The consequence of this long history of erosion has been the loss of some 30–50cm from the surface of the subsoil, the chalk bedrock. Neolithic pits will have been severely truncated, or even eradicated, and any trace of all but the most substantial postholes will surely have gone. The earthworks are universally visible as soilmarks or cropmarks on aerial photographs, with the dark ditches often juxtaposed to traces of "banks" revealed as chalky strips of soil (or parched crops). Upon excavation the bank areas proved, without exception, to be no more than native chalk; protected by the former existence of a bank, they had escaped the worst of the corruption of ploughing and weathering elsewhere. In their solid, brilliant white state they exhibited graphically the scoring induced by ploughing since 1960.

So what remained? The pattern and volume of ditch fill, in combination with the disposition of the bank "shadow" and in the light of experimental earthwork evidence, suggest the ditches were overlooked by a vertically-faced wall of chalk, clearly, and of necessity, reinforced by timber. There were one or two locations where we found postholes associated with this reinforcement frame. At one site the destruction by fire of the timber facade and its collapse en masse into the ditch, preserved signs of an oak framework to which hazel screens (hurdling?) had been attached. Though the ditches were all, more or less, causewayed, it is likely this was due to their initial quarrying function: once enough material had been accumulated for the wall, digging ceased.

Long recorded as a visible earthwork, the 8.5ha causewayed enclosure on the crown of the hill (now known as the main enclosure) was accompanied by two broadly contemporary long barrows, one on its immediate southern flank and the other, much larger, set prominently at the centre of the northern spur (inside the later hillfort). Since excavations in 1960 by Desmond Bonney of the Royal Commission on the Historical Monuments of England, "cross dykes" immediately to the south and east of the main enclosure were also known to be neolithic, as were altogether more substantial outworks on the Shroton Spur some 350m to the east. The true complexity of the earthwork complex, however, was not revealed until an intensive field survey of the entire hill and its surroundings by Rog Palmer began in 1974 and was later joined by a team from the RCHME led by Alastair Oswald.

Earthwork politics

These surveys added an entire series of earthworks on the southern spur of the hill – including a second 1 hectare enclosure (the Stepleton enclosure) and a major outwork system lining the western and southern flank of the hill between the Stepleton and Hanford spurs. In addition, an outwork system was recognised on the western flank of the hill (visible, in the upshot, on air photographs taken by OGS Crawford in 1924!). Finally, in a brilliant piece of detective work, Palmer and Oswald traced the western outwork beneath the massive ramparts of the iron age fort on the northern spur of the hill. In all, the complete outwork system encloses some 60ha.

We conducted a substantial programme of radiocarbon dating, with some 160 carefully-selected samples. Using Bayesian statistics to accommodate all the contextual information, the relative and absolute chronology of this massive complex was successfully resolved. The earliest stage of construction was the main enclosure (dated with 95% confidence to 3680–3630BC) and the southern long barrow, and possibly in addition only the inner east cross dyke (3690–3620). A sequence of four periods of building activity followed, with the Stepleton enclosure and eastward facing Shroton outworks followed eventually by southern and western outworks in period 4, when the main enclosure was still in use (3360–3310BC).

These two, the main enclosure, and the Hanford and western outworks (3510–3320BC), are not only the most substantial constructions, but they also lie at either end of the 3–400 year development of the site. It all seems to add up to a monument, a physically impressive, societally significant, centrally expressive construction, that initially faced east across the Iwerne onto Cranborne Chase and was defended from that direction (Periods 1a–2).

Probably some time between the 37th and 35th centuries BC, the hill came under a series of attacks, including burning. Then between 3400–3300BC the western outwork system was constructed, defending from the opposite direction. The site left the gravitational pull of the southwest and became attached to emergent power centres on Cranborne Chase and to the northeast (periods 3–4). This "turning around" occurred just when developments on Cranborne Chase, as indicated by the vast Dorset Cursus construction, seem to have been coming to a head. It is a unique picture of early neolithic politics in action.

Seasonal gatherings

The very large scale of sampling on the site has allowed us to make credible estimates of the total quantities of material consumed. Distributed over up to four centuries, these are minimal: in one year less than one cow and ten pottery vessels, or somewhat over 1,000 worked and unworked fragments of flint. It seems unlikely that this can represent permanent occupation of any stature. Periodic visits by substantial parties seem to be indicated, and there is some evidence from the sheep remains that these took place in spring, and again separately in late summer.

The vast enclosure system might be taken to require a large population for its construction. But broad-brush calculations reveal that no individual unit is beyond the resources of the likely surplus labour in a total population of around 1,000, over two years in a two month period of agricultural downtime (perhaps the autumn). For many components the time estimate would be much less, but for the main causewayed enclosure and the great southern and western outwork systems this estimate of four months total effort for perhaps 100 people would suffice.

Is this a site at the centre of settlement and social vigour? At every stage of the detailed and fascinating investigations by specialists, the conclusion has been that while the people had come from farming settlements, Hambledon stood on marginal land, still afforested, certainly not intensively farmed – the natural habitat of the red and roe deer, marten and badger whose remains occur on the site. This was border country between populations and geomorphologies, remote from settlement foci.

So the most attractive hypothesis suggests periodic, probably seasonal, visitation by large groups of people. They brought with them selected cattle, pigs and sheep for slaughter and consumption (the first probably as beasts of burden and surplus to the requirements of milk production, evidence for which has been found in pottery residues), and ready-processed grain and other foods. Judged by the imported goods, these visitors came from the north and west, from the river Severn and as far as Devon if not further.

Eight different pottery fabrics represent the breadth of connection from the Jurassic hills to the north to the hard rocks of the south-western peninsula. Non-flint axes came largely from Cornwall, but examples were also located from as far away as the Lake District; two jade axes, probably heirlooms of some antiquity when they arrived, had travelled even further.

These prestigious objects are largely found in the main enclosure, where pits occur with that we would call "rich" deposits, recut or, sometimes, marked by a post. The ditches of this enclosure and its long barrow were repeatedly recut to deposit food debris and objects. Human skeletal material occurs frequently, sometimes with traces of exposure, excarnation and defleshing. Skulls are particularly frequently found in primary deposits in the ditches. Pattern and recurrence are everywhere, an extreme example being the interment of two children, each with a possibly inherited skull deformity, in the same segment of causewayed ditch but in contexts several generations apart.

A rich history

Millennia before the neolithic earthworks were built on Hambledon Hill, native hunter-gatherers had left their mark. We found four pits set on the hill crest that contained pine charcoal, two of them radiocarbon dated between around 7000–8000BC. Elsewhere, similar but rare occurrences of mesolithic pine charcoal, most notably near Stonehenge, have been associated with apparently monumental free-standing posts.

Farming – whether introduced by immigration, or the spread of goods and ideas alone – seems to have arrived in Dorset and Somerset around the turn of the fourth millennium BC. So an agricultural society had been developing in the immediate vicinity of Hambledon for four centuries (the time that separates us from Elizabeth I) before this immense undertaking began; and the total life of the complex was broadly similar.

After the excavation of ditches and the building of walls ceased, strange pit digging occurred in the late neolithic. Makers of Beaker pottery recognised the then defunct enclosures. Early bronze age fields were laid out, and a middle bronze age settlement was built atop one neolithic enclosure. Later bronze age "burnt mound" activity took place in the Iwerne valley, and the tip of the northern spur of the hill was fortified. Early iron age house platforms were dug there within a great multivallate fortification that enclosed the whole northern spur. A Romano-British field system covered the hill, and an Anglo-Saxon cemetery was set by the parish boundary. Oliver Cromwell fought a skirmish there in 1645, and, a century later, Colonel (later General) Wolfe exercised troops before going on to scale the Heights of Abraham.

Before all that, over 5,000 years ago people came in their hundreds to Hambledon Hill, to celebrate rites of passage, perhaps, or renew associations or resolve conflict: but such actions are unlikely to leave any archaeological trace. What we can see are the signs of seasonal gatherings, with the disposal of some dead, much feasting and the dumping of refuse, within areas defined by chalk and timber enclosures and linear outworks. The final completion of the archaeological project that began 35 years ago will give voice to what remains from those activities, and further our understanding of an entire era in British history.

Roger Mercer directed the rescue excavations at Hambledon Hill for English Heritage and its predecessors. Among many who made important contributions to this study are Phil Austin (charred plants), Alex Bayliss (radiocarbon dating), Martin Bell (environment), Mark Copley (pottery residues), Tim Darvill (ceramic clays), Glynis Jones (cereals), Tony Legge (livestock), Jacqueline McKinley (human remains), Alastair Oswald and Rog Palmer (landscape), Fiona Roe (stone artefacts) and the late Isobel Smith (pottery). Hambledon Hill, Dorset, England, by Roger Mercer & Frances Healy, is reviewed in Books.

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