British Archaeology, no 17, September 1996: Features

Understanding how earthworks change

Martin Bell outlines the first results of the country's longest-running archaeological project

Earthworks account for a large proportion of Britain's surviving archaeological monuments, and yet until recently very little was known about how they change over time, how their appearance is altered and how their contents decay. As a result, many of our ideas about, say, Neolithic enclosures, Bronze Age barrows and Anglo-Saxon graves have been based on assumptions about their original state that could - who knows? - prove absolutely wrong.

It was for this reason that the Experimental Earthwork Project was founded 36 years ago. Two earthworks were built, one at Overton Down near Avebury in 1960, designed for comparison with the great prehistoric monuments of the Wessex chalk, and one at Wareham in Dorset in 1963, to aid our understanding of monuments on acid heathland. Artefacts were deposited in both earthworks. Already Britain's longest-running archaeological project, it was designed to last an almost ridiculous 128 years - the earthworks were to be excavated after two, four, eight, 16, 32, 64 and 128 years. Now the results of the first 32 years of research have been published, and the 32-year excavation at Wareham took place this summer (actually in year 33).

It may seem a doddle excavating an archaeological site when you know exactly what has been buried and where. Sometimes, however, it proved surprisingly difficult to locate known objects. One shudders to think what proportion would have been detected if we had not known they were there.

Often at sites such as Anglo-Saxon graves archaeologists find bits of textile or wood preserved in the decay-products of metal - such as a piece of cloth adhering to a brooch - when the rest of the material has gone. This seemed surprising, as it was assumed organic material would disappear long before metal decay-products inhibited the microorganisms of decomposition. The earthworks experiment, however, has shown that textiles often take longer to decay than was thought, with some still surviving after 33 years.

Forensic archaeologists in particular are now looking closely at how the different types of material have decayed over the past three decades, to help them and the police understand what happens to clothing and other items (for instance, leather) associated with present-day buried murder victims.

As for the bones in the earthworks, at Wareham all but one cremated piece had vanished after 33 years. At Overton, although much of the bone remains well-preserved,fungal activity suggests some may not last very much longer. Little wonder that some prehistoric burial mounds produce no sign of human remains.

At Overton, the appearance of the earthwork changed dramatically each year at first, with chalk eroding from the ditch and the bank. After ten years, however, the whole monument had been stabilised by vegetation, and erosion largely ceased. This suggests that, on chalk, all but the largest prehistoric earthworks would have looked striking and 'fresh' for only about a decade. In 20 years, they would have faded into the landscape, and would already have looked like old features. Vegetation has taken longer to stabilise the earthwork at Wareham, with the top of the bank and the edge of the ditch still bare and eroding in places - suggesting that on acid, sandy soils monuments would have looked fresh for longer.

During the period when Overton was eroding, annual winter/summer banding was recorded in the silts washed into the ditch. This was an entirely new discovery, and may establish whether prehistoric ritual activity was seasonal. Ritual deposits are often found in ditches, and analysis of silt banding may provide clues as to the time of year they were buried.

Archaeologists tend to assume that the buried soil in a monument becomes 'fossilized' at the time of burial, and that the seeds, pollen, molluscs and other such evidence in the soil represent a clear record of the ancient environment. We found at Overton, however, that the buried soil had been completely reworked by earthworms, suggesting the stratification of environmental evidence may sometimes be unreliable. At Wareham, by contrast, very little reworking has taken place, and each turf in a stack at the core of the mound can still be distinguished.

One of the big problems for archaeology is that our timescale of knowledge is so short in relation to the time we seek to comprehend. The ambitious timescale of the earthworks project seeks to bridge the gap between the research preoccupations of successive generations. It also highlights the more central role which experimental archaeology ought to play in helping us understand how the archaeological record has formed.

Dr Martin Bell, of the University of Wales, Lampeter, is the Director of the Experimental Earthwork Project for the 1990s. 'The Experimental Earthwork Project 1960-1992', eds M Bell, PJ Fowler and S Hillson, was recently published by the CBA at UKP36.00.

Return to Table of Contents Return to CBA Homepage

The Roman town was neither a failure nor a freak, writes Roger White

Wroxeter, rich in a wealthy land

There may be little to see nowadays at Wroxeter, once the fourth largest town in Roman Britain. Yet the site, which now lies under pasture some miles east of Shrewsbury, is Britain's best-preserved deserted Roman town, and current and recent archaeological work there has shed much new light on the way of life of Roman citizens living at the edge of the Empire.

Wroxeter used to be thought of as a failed town. But excavations by Dr Philip Barker over recent years have shown that its public and private buildings continued in use well into the Dark Ages when greater cities such as London and St Albans had become deserted ruins (see BAN March 1994). Now, geophysical surveys by the Wroxeter Hinterland Project, based at Birmingham University, have demonstrated that Wroxeter was much more densely populated than previously thought. Both housing and the street grid extended right to the town wall, while the northern periphery of the town seems to have been given over to industrial annexes.

And yet, at the heart of the city lies a paradox. Why was Wroxeter so large and important? Where did the wealth come from to build such a place?

In the traditional view, the surrounding countryside in the pre-Roman period was seen as impoverished, especially compared with the sophisticated late Iron Age kingdoms of the South and East. The local tribe, the Cornovii, is usually dismissed as culturally and politically backward. They did build hillforts in profusion and also farmed extensively, siting their farms in ditched enclosures in the river valleys. But they did not mint coins, nor do they appear to have used domestic pottery to any great extent.

The supposed poverty of the Cornovii, however, has always sat at odds with common sense. They inhabited a territory roughly coincident with modern Shropshire and Cheshire, hardly poor counties in agricultural terms. Moreover, the Wroxeter Hinterland Project is showing that the landscape - far from being sparsely populated - was fully settled before the Romans arrived; and the likelihood is that, as is the case now, the bulk of agricultural production in the region was pastoral not arable. Such farming may have brought considerable wealth but may have left few traces in the landscape.

In addition to agricultural riches, the tribe also exploited three out of the four known brine-springs of Britain and had access to copper, lead and silver mines. The Cornovii were certainly rich but with all their trade-routes controlled by other tribes, such wealth may have been difficult to exploit. Presumably wealth was displayed in other ways; in raiding, in building hillforts, or possibly in ritual feasts.

When the Romans did arrive, the tribe was rapidly pacified and, within two generations, the town of Wroxeter was flourishing. The energy that had previously been channelled into constructing hillforts was instead turned to building the urban centre of the administrative capital. We now know that in the surrounding landscape, farms were reorganised to produce market garden crops to feed the army - and then the rapidly swelling population of the town - and it is likely that many of the numerous enclosures discovered through air photography around Wroxeter were founded at this time.

Despite this activity, however, there is little evidence that Romanised lifestyles penetrated deeply into the countryside. Extensive fieldwalking carried out by the Wroxeter Hinterland Project has found that sites with Roman ceramics tend to occur in close proximity to the road network. Isolated farms do not seem to have gone out of their way to acquire the most basic trappings of Roman civilization.

The most spectacular example of this is Whitley Grange villa, which lies in the Rea Brook valley west of Wroxeter. Here excavations in the past few months have uncovered a villa equipped with a bath-house 25m long including a small swimming pool, and another domestic range containing a mosaic 6m square. Despite the spectacular remains, only four coins and less than 100 sherds of pottery have been found during excavation. Clearly, although content to live within Roman buildings, the Cornovian aristocracy did not care to fully equip themselves with all the paraphernalia of Roman life.

Seemingly, the Cornovians were every bit as wealthy as their compatriots in the South East, but like their Iron Age ancestors, they set themselves apart in the manner in which they spent their money and demonstrated their wealth.

Dr Roger White is the Project Archaeologist on the Wroxeter Hinterland Project. Further details about the project can be found on the web pages

Return to Table of Contents Return to CBA Homepage

The old idea of Iron Age warrior societies is fast changing, writes JD Hill

Weaving the strands of a new Iron Age

To many people, the Iron Age in Britain (c 700-0BC) probably seems a period about which we have little left to learn. Traditionally seen as a time of warring 'Celtic' tribes based in hillforts, the period may even seem a little dull compared, say, to revelations about the first Europeans at Boxgrove, the bizarre nature of the Neolithic, or cosmopolitan Roman Britain.

But below this apparently still surface, our views of the Iron Age have seen radical change and controversy in a flourish of new research over the last ten years.

One major cause of this change has been the sheer scale of recent discovery. Iron Age farms and hamlets were densely scattered across the landscape and they are regularly found during rescue excavations. At the same time thousands of new coins and metal objects have been reported by metal detectorists. This mass of new data has inspired new ideas about almost every aspect of the period.

The 'new Iron Age archaeology' is exemplified by work on the Iron Age stone towers - the brochs - of the Scottish Highlands and islands. Recent studies of these brochs include a questioning of their supposedly 'obvious' military function; closer looks at domestic architecture, at how the inhabitants of brochs understood their world, and at their rubbish; and a rigorous debate about whether brochs were the homes of clan chiefs or common farming families. Not all these questions have yet been answered, but the studies all share a healthy scepticism about existing interpretations, and a commitment to testing the received wisdom against the actual evidence from the period.

Of all the new debates on the period, the most acrimonious is over the issue of the Celts. Popular images of the period are usually inhabited by Celts based on contemporary and later written evidence that describes real and mythical peoples who spoke a Celtic language in different parts of Europe over at least 1,500 years. These images have been added to nationalistic, even racist, ideas since the last century to create the conventional modern image of 'the Celts'. This image of the warrior Celts has received great criticism from British archaeologists in recent years (see, for instance, BAN March 1994).

No one is denying that people in Iron Age Britain spoke Celtic languages, or shared certain common cultural traditions with their contemporaries in mainland Europe, such as the use of La Tène 'art'. What has been shown to be untrue, however, is that there existed a single Celtic race whose members all had the same religion, psychological traits, and type of society, and who recognised themselves as 'Celts'. At the heart of this debate is the simple issue of evidence. Should we reconstruct the lives of Iron Age people from the actual archaeological evidence they left behind, or use contemporary Greek and Roman descriptions of other (possibly different) Iron Age societies, or should we draw analogies from later medieval Celtic speaking societies often far removed in time and space? And where the archaeological evidence is clearly at odds with the traditional picture of a Celtic Iron Age, do we ignore it because it does not fit?

The archaeological evidence suggests that, rather than a world of warriors, Iron Age Britain was in fact a more humdrum world of farmers. Probably as much as 99 per cent of the million-plus population were full-time or part-time farmers. Over recent decades, great advances in understanding these farmers - what crops they grew and how, their productivity, and so on - have come from studies of animal bones, plant remains and experimental archaeology.

In the last ten years, however, it has also been recognised that the settlements of these farmers, and the rubbish they threw away, were not quite all that first meets the eye. Their settlements were usually built to physically recreate their mythologies, which they then literally lived inside. Iron Age round houses, for instance, even many farms and hillforts, faced east or south east, not to keep out the wind but to face the rising morning sun. In some cases, round houses were built so that their doors faced where the sun rose on the winter solstice.

Decisions about where other activities took place in and around Iron Age farms were similarly governed by taboos and expectations. Important among these was where 'rubbish' was deposited. Analysis of the pot sherds, animal bones and tools excavated from these settlements suggests many were ritual deposits, and were not just 'ordinary' rubbish casually thrown away. As in many aspects of new research, this conclusion builds on work started by Barry Cunliffe at Danebury hillfort in Hampshire in the 1970s and 80s. Some of this rubbish includes complete animal and human carcasses, and bits of carcasses, both possibly the result of sacrifice. But even broken pot sherds, old tools and worn out quern stones were sometimes placed in special parts of a farm, together with other objects, sometimes even according to a particular order.

Unpicking this complex weave of patterns in the data provides a fuller picture of the beliefs of Iron Age people. For example, in the Iron Age wild animals and plants appear to have formed an insignificant proportion of the total diet. But where wild animal remains are found on sites they often come from ritual deposits. Does this indicate that hunting and eating wild animals and fish was taboo and only allowed on special occasions?

Interpretations of the types of societies these farmers lived in are changing as well. The common image of a Celtic warrior society led by a chief or king may have been the exception rather than the rule. There has been considerable debate about what hillforts were actually for, suggesting their defences were often as much for display or symbol as primarily for defence. At the same time new studies have argued that the evidence from Wessex contradicts the idea that these hillforts were the residences of kings. A more communal and a relatively egalitarian society of small, competitive farming families might be more appropriate for large parts of Iron Age Britain where unequivocal evidence for an aristocracy is hard to find.

This is not to say all Iron Age communities were like this. It is important to remember that the Iron Age lasted for seven centuries of change. Only the last 100 years of the period clearly saw a hierarchical, possibly class-based society with marked differences in wealth and power very different from that found earlier - and only in one part of Britain, namely the South East.

It is also important to stress that the Iron Age was not the same across Britain and Ireland. While different types of chieftainly or kingly societies existed at the end of the Iron Age in south-east England and north-east Ireland, they did not necessarily exist across all of Britain at this time. While there were contacts, and shared cultural elements across Europe, it is the differences in all aspects of life between neighbouring areas that seem to have been more important than the similarities. One region might bury its dead in graves with grave goods, for instance, while next door the people treated their dead in an archaeologically invisible way. One area might have hill-forts and little fine metalwork, while the neighbouring region had the exact opposite. One area might have settlements enclosed by substantial earthworks, while the neighbouring area had open settlements, although presumably both societies were as threatened as each other by aggression.

What these differences appear to show is that the lives, religious practices and types of society of Iron Age people were markedly different, at any one time, in different parts of the country. This is not to suggest these groups had little contact with each other. It is likely that trade and marriages between different groups were common, but that it was very important for groups to mark out their particular differences. The consequence is that it is difficult to talk meaningfully about a single Iron Age Britain.

Perhaps the most important implications of the new studies are for the periods that came before and after the Iron Age. A revised view of the Iron Age has tremendous consequences for how we understand the Romanization of Britain; but equally, there are consequences for the Bronze Age. New work, for instance, has raised the tantalising possibility that iron did not directly replace bronze at the end of the Bronze Age, but rather took over the roles of flint.

This is an exciting time of change in the study of the Iron Age, but the full results of the new research will probably take a few more years to settle out.

Dr JD Hill is a Lecturer in Archaeology at Southampton University, and the author of a recent synthesis of the British Iron Age in the 'Journal of World Prehistory' (1995).

'Reconstructing Iron Age Societies', eds Adam Gwilt and Colin Haselgrove, is a series of essays outlining many of the new ideas on Iron Age archaeology, and will be published by Oxbow in the autumn.Warriors or farmers? This painting of Celtic aristocrats underlines the traditional view, that war came first

Return to the British Archaeology homepage

Return to the CBA homepage

© Council for British Archaeology, 1996