The voice of archaeology in Britain and beyond

Cover of British Archaeology 112

Issue 112

May / Jun 2010


Rare prehistoric finds at major Carlisle dig

Hammerwich hoard "saved" – but who for?

Mary Rose studies query science of tracing migration

Surprising age of new-found stone row on Dartmoor

in brief & phase 2


University archaeology

THE BIG DIG: Discovering Bosworth

The Buried Gods of Gogmagog

Three Men and a (Leaky) Boat

on the web

Caroline Wickham-Jones investigates the archaeological value of encyclopaedic websites and Stuart Jeffrey describes the Grey Lierature Library


A child's gift to science, the human remains debate


Your views and responses

CBA Correspondent

Mike Heyworth explores the great rewards and challenges of undersea archaeology


ISSN 1357-4442

Editor Mike Pitts


THE BIG DIG: Discovering Bosworth

"Up with my tent there!" says Shakespeare's Richard III on the eve of the Battle of Bosworth. But where exactly was that? Memory failed, but after 525 years archaeology has located the battlefield that ended the War of the Roses. Glenn Foard reports.

Richard III Glass

Richard III in a stained glass window in Penrith

On August 22 1485 the last English king of the House of York, Richard III, was killed on the battlefield of Redemore in Leicestershire, defending his throne against the pretender Henry Tudor. Within hours the crown, supposedly found in a thornbush, was placed upon Henry's head and a new royal dynasty was born.

Despite being such an iconic moment in English history, the memory of it became obscured and the supposed site of the battle migrated across the landscape. In 1985 historian Colin Richmond pointed out – to widespread controversy – that Ambion Hill, said to be the battlefield since at least the later 18th century, seemed not to fit the original accounts. This set off a quarter century of dispute between historians, with alternative sites scattered across more than 6km between Market Bosworth and Atherstone.

When Leicestershire County Council were planning to revamp their Bosworth Battlefield Visitor Centre (opened in 1974 on Ambion Farm, which they had bought the year before, and have recently renamed Bosworth Battlefield Heritage Centre), they turned to archaeology to resolve the problem. In 2005 the Battlefields Trust was commissioned to undertake a study over three seasons, with £154,000 from the Heritage Lottery Fund. An interdisciplinary team was brought together, with historians, palaeo-environmentalists, a soil scientist and a placename scholar, working alongside archaeologists and a group of volunteers. We systematically applied each step of the evolving methodology of battlefield study.

Bosworth proved a remarkably difficult challenge, and in the end it took three and a half years to find the site and a further year to broadly define its extent and character. Like most late medieval English battles, Bosworth is poorly served by primary accounts. After reviewing the main secondary and antiquarian works to establish current knowledge, we reworked all the evidence from the original sources, placing it in a wider historical context and securely grounding the few historical "facts".

This showed the site near Atherstone (suggested by Michael Jones in 2002) was improbable. Placename, map and early 16th century sources clearly locate the battle in Dadlington or one of the four adjacent townships, which include Ambion Hill. A handful of primary topographical clues include the battle names Redemore, Brown Heath and Bosworth Heath (already discussed by Peter Foss in The Field of Redemore, Kairos 1998); a reference in an early ballad to the Duke of Norfolk being killed beside a windmill; and the early 16th century historian Vergil's description of Henry's army manoeuvring around a marsh to protect his right flank, so putting the sun at his back, in order to attack the right flank of Richard's army. Find the marsh and one had probably located the battlefield, as wetlands were not common in medieval Leicestershire.

Richard III Plaque

Memorial plaque at the traditional battle site – 3km from the actual field.

We digitally mapped the early modern landscape and placenames from the historic maps. We reconstructed the medieval landscape of open field furlong, meadow and pasture, from field survey and aerial data for headlands, slades and ridge and furrow. Finally we looked at early modern and medieval written landscape sources. The physical geography was defined using the NEXTMap Britain digital terrain model; by reworking and extending the soil survey data to create a new, more extensive 1:25,000 soil map; and refining the drainage pattern with reference to the early maps and relevant earthwork and soilmark evidence. The marsh could only have lain in the area identified as open field meadow – a zone which proved to closely mirror the distribution of alluvial soils.

Ambion Hill could not be the battlefield because it lay in open field furlongs. Instead, placename evidence drew us to low-lying ground at the junction of four of the townships, where fen-related names concentrated. There were even adjacent heath and moor names which linked to other names for the battle. The eastern part of this zone was where Foss had argued the battlefield lay. A terrier was then identified – a church-owned property document – which showed that the adjacent hill, know as Crown Hill by 1605 and the site where according to local tradition Henry Tudor was crowned, was known as Garbrodys in the 1470s. We even found a reference to Redemore dyke in the 1530s, to complement the 13th century occurrence discussed by Foss, but frustratingly this key name remained unlocated – somewhere in or abutting Dadlington township, whose chapel had housed the chantry to the battle dead. Further complications arose when Dadlington proved to have had detached rights in several adjacent townships, including on the heath in Higham on the Hill.

During the initial field inspection an area of peat was identified, in Fen Meadow, and so a systematic augering survey of the fen zone was undertaken to locate, characterise and date any peat deposits. No others were found, but the Fen Meadow feature proved substantial, more than 100m across, with pollen analysis suggesting a medieval wetland. It looked as though we had located themarsh exactly where Foss had placed it. However radiocarbon dating finally demonstrated it had disappeared many centuries before the battle. Given the strength of support for the Foss site, we commissioned a re-analysis of the peat deposits but with the same answer.

By now a local farmer had reported another peat deposit, in the area once called Fen Hole, missed in the augering survey. Though seeming too small to be tactically significant the feature gave a medieval date. It seems likely that we still do not fully understand the extent and character of the medieval fen, in part perhaps because of decay of peat deposits through desiccation and the impact of cultivation, for the marsh on the battlefield is recorded as having been drained before the mid 16th century. The peat in Fen Hole may be just the last surviving evidence of a more extensive though still small area of marshy ground extending northward along this stream course and its tiny western tributaries.

Before embarking on time-consuming survey for metal artefacts from the battle, ideally one securely locates the battlefield using other evidence and re-interprets the battle accounts in the context of the reconstructed terrain to create a hypothesis as to how the action played out. This is then tested and refined through carefully targeted intensive sampling for battle archaeology – within and outside the presumed site – seeking a pattern of metal artefacts which relates to the action. In this way the battle archaeology provides data wholly independent from the documentary sources. At Bosworth, though we were able to dismiss Ambion at an early stage, the subsequent difficulties with terrain reconstruction meant metal detecting began in reconnaissance mode – sampling using 10m spaced transects, purely for non-ferrous artefacts – to cover a wide tract of landscape in search of a concentration of battle archaeology.

For a 17th century battlefield this would be a relatively simple task because of the large-scale use of lead bullets, which survive well and are easily recovered. In contrast, artefact scatters on a medieval battlefield appear normally to be far more ephemeral, comprising mainly small numbers of fitments from personal and horse equipment. This makes the battle archaeology difficult to separate from the background noise of metal artefacts left by centuries of agricultural and domestic use. Disentangling a meaningful pattern for any period requires consistent, systematic sampling with careful recording of the speed and intensity of work by each detectorist in the team (using GPS tracking) as well as individual recording of each artefact recovered. The low density of artefacts for a medieval battle means that a reconnaissance survey could simply miss the site.

The apparent strength of evidence for Foss's site meant that for three seasons our effort focussed primarily on providing consistent coverage between Moorey Leys, Dadlington and the Sence brook, though we also sampled potential areas elsewhere in the five townships. Only in the main zone did we recover any probable battle-related finds and then with only two slight groupings. That from close by the site of Dadlington windmill suggested this might be the mill where the ballad located the Duke of Norfolk's death, but intensive resurvey failed to yield a meaningful wider pattern.

During a six month extension to the project, with a medieval marsh in Fen Meadow disproved and that in Fen Hole confirmed, we shifted the focus of survey. In the final week, as we started to close the last gap between previously surveyed areas on the western periphery, we recovered a single 30mm lead ball. Such larger lead munitions have such a specialised function that we knew immediately we had found the main action. Now we could move to more intensive survey, at 2.5m transects, increasing the coverage from under 20% to approaching 80% of the field surface – though we stayed in non-ferrous mode because first we needed to define the overall extent and structure of the scatter.

After almost a full season of further work we now have (as the printed magazine went to press) 30 lead munitions – far more than from all the other 15th and 16th century battlefields in Europe put together. In a distinctive spatial association with these is a light scatter of other objects. Though in no way comparable to the dense scatter of buckles, strap-ends and other such artefacts seen at Towton (Yorkshire, 1461), which now seems to be quite exceptional, the Bosworth scatter contains finds of such high status and specific function that they must be from the battle. In earlier seasons we had worked around three sides of the battlefield, to within a few metres of the scatter, but recovered no sign of the battle – just the sparse isolated scatter further east which now appears to be from the rout and destruction of Richard's army.

Finally, and as if to quell any remaining doubters, we have recovered perhaps the most iconic of artefacts that one could have wished to find on Bosworth Field: a silver-gilt boar. Its location, beside the site of Fen Hole, suggests it has the most powerful story to tell. The boar was Richard III's own badge, given in large numbers to his supporters. While most of the badges were of base metal this one is silver-gilt, and thus would have been given to a knight or someone of higher status. This man was probably a knight in the king's retinue, who would have ridden with Richard in his desperate cavalry charge, launched as victory was falling from his grasp, in a last bid to kill Henry. He came so close, but then his small body of knights were driven back and the king himself cut down as his horse became stuck in a mire. Does this badge identify that very location?

The battlefield lies a kilometre to the west of the site suggested by Foss and 3km from Ambion Hill in a location never before suggested. The battle archaeology fits neatly in many ways with the reconstructed terrain and the battle accounts, and straddles the Roman road along which the armies probably advanced. But many uncertainties remain, not least because the battlefield-wide survey is incomplete due to the restrictions of the agricultural regime. Neither have we undertaken intensive sampling in all-metal mode in the core area, to seek the ferrous arrowheads from the initial arrowstorm. If they survive such a scatter may provide information as to exactly where and in what orientation the bodies of infantry first clashed. And this would provide some evidence by which to focus a search for the mass graves, for which we currently have no evidence.

While the evidence for the battle itself is important, the study has wider implications. The discoveries have generated a new strand of archaeologically-led research into early gunpowder weapons. We are bringing together specialists in modern ballistics, in early gunpowder weapons, and in the application of nondestructive scientific techniques to the study of medieval and early modern metal artefacts.

With the latter the aim is to characterise the lead munitions and the evidence they contain using techniques such as 3D laser scanning to provide a complete surface model, and neutron tomography to provide 3d images of their internal structure. Lead munitions for the smaller field artillery in the late 15th and 16th centuries included a wide variety of types. Some were of solid lead, others were cast with a roughly square "dice" of iron in them, or with a stone pebble or flint, or even a cast iron ball. Only the latter type is so far missing from Bosworth. Why this was done is unclear but the first step towards an answer is to accurately record the structures of the munitions.

We have also begun a study of the internal characteristics of barrels of original guns in European museums. We are seeking to manufacture gunpowder to an early recipe so we can reproduce the ballistic properties of the guns in experimental firing using reproduction barrels and roundshot. This should help us further understand the firing and impact evidence on the original munitions. It should also allow us to consider gun positions on the battlefield once the final range – after bounce and roll – of different types of gun has been established. The principles should have wide implications for the study of early firearms on battlefields across Europe, for Bosworth will not be unique in its evidence for these weapons.

Research should not focus purely on the gunpowder weapons. We need an understanding of what pieces of medieval equipment are likely to have been broken and lost by troops of different status in the action or during the later stripping of bodies, in order to interpret the other metal small finds scattered across the field. It will also be important to consider factors determining the survival of metal artefacts; the likely current trajectory and speed of their decay; and how best the remaining battle archaeology can be conserved in situ. This demands an understanding of the influences of soil chemistry, and both recent and long-term agricultural use, including the impact of modern agrochemicals and new cultivation machinery such as power harrows. Only then may we be able to design effective strategies to mitigate the threats.

As the Bosworth project draws to an end, in April 2010, it is clear that the battlefield still has a complex story to tell – one we have only just begun to unlock. Because it remained undiscovered for so long, it has almost completely escaped the impact of treasure hunters. Moreover the distribution of lead munitions provides a clear structure within which to seek a patterning of other artefacts. Thus Bosworth offers the potential to understand more fully than ever before the archaeology of a typical late medieval battle – increasingly Towton appears to be exceptional in the scale and character of its battle archaeology. But Bosworth also shows warfare in transition, with gunpowder weapons just beginning to transform the battlefield, an important research theme for battlefield archaeology.

I would like to thank the metal detecting team and other volunteers and advisors; Anne Curry and Janet Dickinson; David Hall; Tracey Partida; Mark Page; Rodney Burton; Barrie Cox; Rob Janaway; Graeme Swindals; Ben Geary and others of Birmingham Archaeology; Richard Knox; Derek Allsop; Evelyne Godfrey; Richard Mackinder; University of Leeds; Cranfield University; Royal Armouries. Glenn Foard is project officer with The Battlefields Trust.

<< prev | index | next >>

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