The voice of archaeology in Britain and beyond

Cover of British Archaeology 101

Issue 101

July / Aug 2008



Early Scottish gardens unturfed

Axes could be 0.5m years old

In the press

Listing lobby was no hot air

Kent Anglo-Saxon cemetery could be royal

Poetry to assist transfer of Hadrian's Wall collection

In Brief & Phase 2


The Copper Age
Special: did Britain have a fourth age?

Portable Antiquities
The Scheme must go on

Drawing Stonehenge
A major fieldwork project is explored by six artists

Severn estuary
Martin Bell describes the unique world of ancient mudflats

Gin Drinker's Line
Insights into WWII Hong Kong defence system

A Professional Mockery
Gary Lock on difficulties in obtaining "grey lierature"

on the web

Recommended websites
Caroline Wickham-Jones explores the realms of archaeological fiction and a look at Barwick-in-Elmet Historical Society's website


your views and responses

CBA correspondent

Campaigns, comment and communications from the CBA
Mike Heyworth says it's time to think big – The CBA at Discover Archaeology LIVE, London Olympia


An exhibition to make you think (and a bog body)

Mick's travels & more travels

Mick Aston goes to Glamorgan in search of monasteries, and Jon Cannon tours south-east Wales

in view

New columnist Greg Bailey probes a coming major TV series – BBC's Bonekickers

my archaeology - NEW!

Neil MacGregor: The accidental archaeologist and new director of the British Museum


ISSN 1357-4442

Editor Mike Pitts


The Severn Estuary: addressing the ecological footprint

Last autumn the government launched a study of the Severn barrage. It would reach 16km between Cardiff and Somerset, and generate 5% of UK electricity. Yet this contribution to reducing carbon emissions was condemned by environmental groups. As Martin Bell explains, besides important wetlands the Severn estuary has very special evidence for the daily lives of prehistoric people.

As the government struggles to achieve its renewable energy targets, the Severn estuary is once again in the public eye with revived plans for a tidal barrage. The area is of high nature conservation status. It is a site of special scientific interest, a special area of conservation and a Ramsar site (defined by an inter-governmental treaty signed in Iran): it is a wetland of international significance. There is widespread concern about the possible effects of a barrage on bird and fish populations. Not so widely appreciated, despite 25 years of intensive archaeological work, is the importance of the Severn estuary as a heritage resource.

Ancient sites are deeply buried within a thick sequence of 10–15m of postglacial sediment. Much of the wetland has been drained and protected by sea walls over the last 2,000 years to create agricultural landscapes. Many of the recent archaeological discoveries are to seaward of the present wall, in the vast intertidal area created by the second highest tidal range in the world, 14.8m at spring tides. These are the best times for archaeologists: low sites are only exposed for some two hours at spring tides, so the working environment is challenging. The moving water cuts sections through the sediment sequence, revealing outstandingly preserved archaeological sites. As sea level rises, such erosion is accelerating.

Discoveries first came about through the activities of an amateur archaeologist

Discoveries first came about particularly through the activities of an amateur archaeologist, the late Derek Upton. In his regular walks across the intertidal zone he found bones, pottery and the remains of wood structures. He recognised the significance of these, not because of any academic training, but as a result of his intimate knowledge of the estuary, its changing environments and past ways of life. Concurrently the geologist John Allen was examining the Severn estuary's sedimentary sequence with detailed field investigations. These established beyond doubt that many of the discoveries which Upton had made were stratified in prehistoric deposits.


Around 9000BC, in the early postglacial period, there was no estuary, just a wide embayment stretching from the north Devon coast to Pembrokeshire into which the river Severn flowed. As sea level rose with climate warming after the end of the ice age, the Severn valley was progressively flooded: first the deeply incised part of the channel, and then around 6000BC the great broad sweep of the valley to create the estuary we see today. The rising sea drowned a great forest of oak trees with tall straight trunks which showed that they had grown in dense climax forest, quite unlike the low branched park oaks with which we are familiar today. As this woodland was drowned, so higher ground on the Welsh shore at Goldcliff became an island; and on the edges of this is a complex of mesolithic sites, which we excavated between 1992 and 2003.

Finds threw light on the social use of space unconsidered until Romans introduced latrines

Sediments accumulating around the former island buried a sequence of distinct activity areas left by hunter-gatherers between 6000 and 4800BC. At two of these areas people cooked fish, particularly eels; at another they butchered giant extinct cattle (the aurochs) and used heated stones during cooking. On three sites tentative evidence was identified for small tepee-like shelters. An idea of what such structures may have been like is provided by a partial reconstruction attempted at Butser Experimental Farm, Hampshire during a mesolithic weekend.

The waterlogged sedimentary sequence preserved a far wider range of evidence than generally encountered on mesolithic sites, including some wood tools. Petra Dark identified the earliest evidence for human intestinal parasites, showing that there were defecation places on the margins of activity areas. That illuminates an aspect of the human social use of space which is virtually unconsidered in British archaeology until the Romans introduced latrines.

What furthermake this landscape so exceptional, are human and animal footprints in laminated silts above and around the mesolithic island edge occupation sites. The very best footprint was found in 2004 after the conclusion of our excavations, when we returned to film for the first BBC series of Coast. By chance, erosion by the sea had uncovered an exceptionally well-preserved print of an 11-year-old. Research by Rachel Scales has shown that the majority of the human footprints are from children, some as small as three or four: the young had an active role in mesolithic society, with all segments of the population engaged in coastal activities.

Sedimentary and pollen research by John Allen and Petra Dark have shown that the laminated silts with footprints show evidence of annual banding; coarser sediments were deposited in winter and fine-grained sediments in summer. The best preserved footprints are in fine-grained sediments and clearly represent activity at the height of summer. Along with human footprints there were also many red deer, some aurochs and wolf. Bird footprints are numerous, and of particular note are those of the crane, no longer native to Britain but clearly breeding here in large numbers in the mesolithic, since its footprints are abundant in the finest high summer sediments. Interestingly the crane now seems set for reintroduction with tiny pioneer colonies established in Norfolk and at Slimbridge on the Gloucestershire coast of the Severn estuary.

Petra Dark and Alex Brown have identified much evidence for burning preserved in the Severn sediments, in some, but not all, cases directly associated with mesolithic activity horizons. At other sites such as Redwick there is evidence of burning but no known settlement. Large trees in the now submerged forests were burnt, and also reedswamp.

This discovery complements evidence for mesolithic burning in the uplands, where fire was said to create open areas attractive to grazers, particularly red deer. That seems unconvincing in the highly dynamic environment of the Severn estuary where the environmental record shows a constantly changing mosaic of different vegetation communities which would have naturally provided abundant browse for grazers. In this area the encouragement of plant resources especially favoured by people puts this hypothesis in doubt. The density of artefacts is insufficient to infer sedentism; discrete activity areas are too clearly defined (implying brief and not overlapping events). Nor, however, do these discoveries support another traditional model, that of bipartite winter coast/summer upland activity. The wide range of evidence – plant remains, animals, footprints, sediments and burning – all indicate visits to the coast at various times of year, with activity mostly concentrated in mid to late summer and autumn, but some short-term activity in spring and, on one site only, during winter.

Later activities

Current fieldwork by myself and Alex Brown concerns a complex of former stream channels at Peterstone, where there are signs of activity by farming communities in late neolithic/Beaker times (2500–2200BC) and during the middle bronze age (1500–1150BC), long after the hunter-gatherers had gone. We have identified three distinct types of wood structure in the channels. There are alignments of split oak posts along the edges which are of late neolithic/Beaker date; settings of between one and three large wood posts in channel centres; and two hurdles dated to themiddle bronze age.

Some of the structures may relate to fishing or landing places for boats. Discovery of the wood handle of a palstave-type bronze axe was reported in British Archaeology (News, Sep 2006). The head had been hacked off with an axe and buried in an arrangement of stakes, which suggests ritual deposition. Woodchips, bones, heat-fractured stones and pottery seem to have been swept into the channels from adjacent occupation surfaces; however, the occupation site itself may have been eroded away.

In the middle bronze age much more intensive activity occurs, with many occupation scatters and wood structures. At Redwick we have totally excavated a group of four rectangular buildings. Some have possible internal sub-divisions, perhaps animal stalls, and there are many cattle footprints around the buildings, together with a few human footprints, including those of a child aged five to six. Scatters of artefacts outside entrances suggest the buildings may have housed people as well as cattle, but if so artefact numbers suggest visits were of short duration.

At Goldcliff we excavated a group of eight rectangular iron age buildings (500BC–AD100) in the early 1980s. One of these had animal stalls, and cattle head-biting lice and cattle footprints were also present. Here 17 iron age wood trackways were found, some leading directly from buildings seawards. This highlights the role of boat transport, also demonstrated by discoveries of carefully worked planks from sewn boats in bronze age contexts at both Goldcliff and Caldicot.

Seasonality evidence from plants and animals at bronze age Redwick and iron age Goldcliff points to activity mainly in spring and summer, the time when dairying would have been important: rich saltmarsh pasture is likely to have been a valued seasonal resource. In addition to the seasonal activity within the wetlands, the estuarine resources were also exploited by permanent settlements at the wetland edge. The most fully investigated is Brean Down, across the water on the Somerset coast, which I excavated in the 1980s. Here, in contrast to the peatland sites, there were round buildings, and abundant artefacts including evidence of high status gold bracelets, acquired may be through exchange for salt which this bronze age community was extracting from sea water.

It remains a matter of debate whether the rectangular structures and impoverished material culture of the Welsh wetland sites reflect a seasonal dimension to communities living in permanent settlements similar to those at Brean. Alternatively there may be a cultural divide across the estuary. Another possibility is that in the wetlands we may have lower status, maybe somewhat marginalised communities or sections of the population, focused more on wetland resources.

The bronze and iron age pattern of seasonal pastoralist activity was attuned to the natural rhythms and cycles of the estuary. Land use changed dramatically from the Romano-British period onwards. Iron age marine incursion and saltmarsh gave way to drier conditions. Romano-British communities dug ditches and constructed low banks to reduce tidal inundation. Henceforth the Severn Levels could be used for a much wider range of activities including year-round settlement and sometimes crop growing. In this way people insulated themselves from the cyclical changes of the estuarine environment. By so doing, however, they made themselves more vulnerable to rare extreme storm events. One such is historically recorded in AD1607 and led to devastating loss of life; some have interpreted this as the product of a tsunami.

The Roman and medieval drainage and landscape history of the levels has been intensively researched by Stephen Rippon of Exeter University, who established the distinctive character of the levels' environment and their early origins. The archaeological importance of the reclaimed levels is further highlighted by the discovery of the Barland's Farm Romano-British boat, excavated by Nigel Nayling and dated to the late third or early fourth century AD. The Magor Pill medieval boat which dated to AD1240 was exposed in the intertidal zone and also excavated by Nayling. He also, with Kate Howells and Kate Hunter, accomplished the truly heroic feat of recovering and recording the mid-15th century AD ship discovered whilst building an arts centre on the waterfront at Newport.

A sustainable future?

If the Severn barrage goes ahead its effects will undoubtedly be as significant for the heritage as for wildlife. The intertidal sites do not just represent better-preserved versions of the sites we have on dry land, they provide a different perspective on prehistory: the footprint evidence for the role of children, the evidence for defecation practice, the use of plant resources and the role of burning. They show that seasonal movements associated with cattle husbandry in the bronze age and iron age are more significant than generally assumed, and were associated with structural forms contrasting with those on dry land.

A barrage will reduce the tidal range, permanently submerging the lowest palaeolithic and mesolithic sites. The effect will be much more serious for higher sites, such as that currently under investigation at Peterstone. These will dry out, and the wood and organic evidence will be lost. Barrage construction will increase erosion in some areas and sediment deposition in others, burying some sites and exposing others to destruction.

Of equal concern to the effects on the intertidal zone are those of the associated infrastructure developments behind the sea wall, and the possibility of access roads and other facilities crossing the historic landscapes of the levels. Experience also shows that one development tends to stimulate others – just as building the Second Severn Crossing stimulated substantial commercial development in Avonmouth and on the fringes of Newport and Cardiff. The probability is that a barrage would also be associated with substantial "green developments" designed to compensate for the loss of some key wildlife habitats.

We have experience of this in the Severn, where a barrage across Cardiff Bay provided the centrepiece for a major urban regeneration project, the stunning futuristic character of which will be familiar to viewers of the BBC television series Torchwood. Permanent impounding of water behind that barrage led to a loss of tidally exposed mud, important as bird feeding grounds. To compensate, a huge nature reserve was created along the levels immediately landward of the major intertidal complex of mesolithic to iron age archaeological sites at Goldcliff. Here lagoons were dug which are flooded by the sea at spring tides and provide alternative bird feeding grounds. That created a nature reserve, the Newport Wetlands, which quickly achieved its target bird populations.

However, that achievement involved a very delicate balance between nature and heritage conservation. Assessment by Glamorgan-Gwent Archaeological Trust had shown that 0.8m below the present surface was a buried Romano-British landscape of drainage ditches and seabanks. The lagoons had to be shallow and carefully designed to minimise damage to the Romano-British landscape and themore deeply buried prehistoric levels.

The delicate balance between development pressures, nature conservation and heritage represented by this example is but a foretaste of the far greater challenges which we now face. What will the concept of sustainability actually mean for the future management of the Severn estuary? Will its contribution be to energy needs or alternatively to nature conservation, biodiversity and heritage? What is far from clear is whether in this case an ideal balance between energy and nature can be achieved.

Martin Bell is professor of archaeological science at Reading University. The Severn estuary research is funded by Cadw, the Natural Environment Research Council (NERC) and the British Academy. See Prehistoric Coastal Communities: the Mesolithic in Western Britain, by M Bell (CBA 2007), reviewed Books May/Jun

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