A Guide to the Archaeology of Conflict

© T.L.Sutherland 2005

Identification of sites of conflict

The exact locations of most British battlefields have not been archaeologically proven. However, the presumed sites of several battlefields are currently undergoing reevaluation by various specialist bodies. Possibly the most famous of these is that of the Battle of Bosworth Field, where King Richard III was defeated by of Henry Tudor, who subsequently became King Henry VII. This suggests that there is potential to uncover evidence of conflict across larger areas than those currently defined by documents such as the English Heritage register of Historic Battlefields (English Heritage 1995).

Figure 10:

The location of Towton Hall, and the battlefield (BF), which lies between the villages of Towton and Saxton

At Towton, for example, a mass grave of battle victims was found in the centre of a village, over a mile from the recognised centre of the battlefield (AD 1461) (Figure 1). Unfortunately, this area is not officially recognised as part of the battlefield (Sutherland 2000a).

Evidence of conflict

The quantity of available historical and archaeological evidence for a conflict may vary extensively. In Sweden, for example, the skeletons of the Gotland soldiers who fought against the Danes in the battle of Wisby (1361) have been found in mass graves still wearing their medieval armour and mail (Thordeman, 1940) (Battle of Wisby Swedish web site).

In Britain, on the other hand, even the approximate location of many battle sites, for example the battle in which Boudicca, leader of the Iceni tribe, fought and was beaten by the Roman army in AD 61, are unknown.

Generally speaking, the earlier the period, the more difficult it is to find evidence associated with a historically documented battle. In specific terms, however, evidence of conflict is available from all periods from the Neolithic (as noted above) to the present day. At Talheim Germany for example, a mass grave provided evidence of a number of a group of individuals who had been attacked by aggressors using Neolithic stone axes (Wahl, und König, 1987).

How large might a battlefield be ?

The word ‘battlefield’ suggests a sizeable and extensive piece of ground. In the Register of Historic Battlefields English Heritage describe the difference between ‘a battle “site” and a battle “field”, only the latter category being susceptible to the drawing of a boundary line’ (Brown 1998, 2).

The terminology ‘site of conflict’ however, eliminates the suggestion that the area has to be extensive. The siege of a castle, for example, is not classed as a typical battlefield, but might still contain similar evidence to that of a medieval battle, with the addition of the structural evidence which would have been fought over. Such a site would be relatively simple to define, as the evidence of the structure could be used as the focal point of the engagement. Battles on open ground, however, need some other criteria upon which to anchor the conflict and this by its very nature is rarely easily available. It is therefore important to be able to identify the types of terrain that were typically favoured in each period. The Roman army, for example, generally chose open ground on which they could use their cavalry tactics (FLAVIUS VEGETIUS RENATUS). This knowledge permitted Arminius, the leader of the Germanic tribes who had been educated in Roman military tactics by the Romans themselves, to ambush and defeat the Roman army led by Varus along a narrow defile in the Teutoburger Forest in AD9 (Wilbers-Rost 2004).

In order to estimate the potential size or extent of a scene of conflict it would, ideally, be necessary to involve an expert in the period in question. Military historians have a great deal of knowledge of particular battles but this should be used in conjunction with battlefield archaeologists who can use the historical information and any available archaeological data to interpret the most likely scenario. This highlights the multidisciplinary nature required to successfully analyse the archaeology of battle.

Different types of battlefield sites

Conflicts have been fought upon a wide variety of sites. These include examples in towns and cities, such as the medieval battles in Tewkesbury, England (1471) and St. Albans, England (1455) which had minimal impact on the settlements themselves; the Boudicca revolt in Colchester (AD61) which totally destroyed the city; and more recently in cities such as Berlin, Germany (1945) or Baghdad (2002). Other conflicts took place across open ground such as Towton (1461) or expanses of moorland as at Culloden (1746). Some battles such as Stamford Bridge and Fulford (both 1066), or the Battle of the Boyne (1690), near Dublin, Southern Ireland, were fought by or across rivers or watercourses. Other battles such as Maldon, Essex, England (991) have been fought on coastlines, whereas evacuations or invasions such as those at Dunkirk (1940) and Normandy (1944) were fought both on land and at sea. Many naval battles took place in relatively shallow waters leaving archaeological remains of the event. For example, the remains of Henry VIII’s warship, the Mary Rose, was discovered and raised from the Solent estuary outside Portsmouth. This evidence has allowed archaeologists and historians to assess how such a fighting ship was originally built, how it was subsequently modified and how it was equipped for battle. Significantly, the stores of Tudor longbows found within the hull of the ship have allowed scientists to substantiate historical documents which stated that bows with draw weights of over 150lb existed and were intended to be used in battle (Stirland 2000, 124). Although the site of the Mary Rose might not initially be perceived of as a battle site, the fact that it was engaged in conflict when it sank means that it was carrying men and artefacts used in combat. This information can therefore be used to investigate contemporary sites on land. The site of the ship itself is, however, comparable to that of a siege, where the related information is generally confined to within a limited area. The site of the battle, should it ever be investigated, consisting of fired cannon balls, and objects lost overboard from other vessels within the engagement, would cover a much larger area. The bed of the Solent estuary is therefore the repository for the evidence of the battle, of which the Mary Rose is just a part.

In a similar manner, wrecks of the HMS Hood and the German battleship, the Bismarck, represented a fraction of the remains of the Second World War battle of the Atlantic. Just as it was impossible to perceive, during the Second World War that the images of these sunken ships would one day be available for analysis, as is now the case, so it was once inconceivable that evidence of ancient battles would be analysed in great detail. This highlights the variable and often unpredictable nature of analysing sites of conflict. It is therefore possible to find evidence for conflict even in unexpected locations.

As the sites of many famous battles cannot be identified with any great accuracy, then archaeological evidence for these conflicts could be found in the future, even by chance. An example is that of Mons Grapius (AD84), fought between the Roman general Agricola and the British chieftain Calgacus, which has not been more accurately located, than being somewhere in Scotland.

A successful example of such a discovery has been cited above, when Major Tony Clunn’s metal detector search led to the discovery of, what has been interpreted as evidence of the Varus (AD9) battlefield north of Osnabrück, Germany. However, evidence of Roman militaria is also found on British sites that are not, as yet, perceived of as being Roman battlefields. For example, the discovery of what appear to be fragments of Roman military armour and weapons on an archaeological site near York could mark the location of a military engagement (Sutherland forthcoming 2). Only a detailed analysis of the evidence and the context from which it derived, will allow an accurate interpretation of such evidence.


© 2005 Tim L. Sutherland & Simon H. Sutherland