This study [of bullet scatters across Edgehill's battlefield] undertaken for the Battlefields Trust between 2004 and 2007, represents the most extensive systematic archaeological investigation of a 17th-century battlefield so far undertaken in England.
The modern archaeological study of the battlefield, along with a reassessment of historical records and historic terrain research, all combined to form Glenn Foard's 2009 reinterpretation of the Edgehill battlefield. Most notably, the research and archaeology concerning the battle's case/hail shot finds were instrumental in the reorientation of the deployments and the placements of the parliamentarian army. Unlike earlier battles, where archaeology has been utilised to successfully secure battlefield locations, the ability with this battle to also seemingly establish the orientation, munitions boundaries of individual frontages and approximate deployment lines of the defensive army is revelatory. It might be argued that historic terrain reconstruction and a strict interpretation of topographical details from the primary accounts alone was always sufficient to dramatically reinterpret the Edgehill battlefield, but the additional wealth of archaeological detail also supports Foard's battlefield suggestions, while intricate detail from the finds produced from the survey have also provided valuable insight into the battle itself. While the conditions and circumstances at Edgehill were mostly ideal the methodology and techniques developed and exercised during this survey proved their potential to provide remarkable results (and provide lessons for the subsequent successful survey and search for Bosworth's battlefield).
Dr Glenn Foard is a leading authority on battle archaeology: Following is only a small selection of interesting, although isolated, extracts for the general reader from Foard's observations relating directly to Edgehill, the survey and its battle archaeology (in chronological order).
We have already seen that the reconstruction of the historic terrain suggests need for revision in the location and width of frontage of the two armies [when considering popular 20th century - and similar - conjectures]. The battle archaeology can be seen to provide strong supporting evidence for this reinterpretation, once one takes into account the gaps in the pattern, caused mainly by the destruction resulting from construction from the 1940s onwards of silos and railways of the MOD munitions depot. […]
Although we have too few bullets at present, there are initial clues. For example, two groups of case shot may suggest the location of artillery pieces.
Comparison of written and archaeological records, complemented by experimental work, may bring advances in the methodology of recording and analysis. It may enable assessment of the effectiveness of particular survey techniques, or reveal archaeological signatures of different types of action, as at Edgehill where recognition of the distributions of bullets fired as case shot from artillery indicates the potential to reconstruct the exact placement of battalions in a battle array.
Systematic monitoring of data collection at Edgehill has demonstrated that on that site with an experienced team an intensity of survey of 10m transects with a reconnaissance speed of c.8-12 metres per minute was only just adequate to identify case shot locations, and that 2.5m transects were the minimum for tracing the orientation of the case shot scatters.
There is need for: […] a case study on a well-preserved battlefield fully to explore the potential of bullet scatters, including particular aspects such as case shot scatters and firing lines; the Edgehill study suggests that such aspects may enable the recognition of individual battalions
Urban development, road construction and mineral extraction will destroy or occasionally mask the archaeology in the areas upon which they impact. Large scale earthmoving can transform the detail of relief and drainage. Such activity will also threaten other surviving aspects of terrain. Stripping and redeposition of topsoil will redistribute artefacts and so destroy the detail of spatial patterning. Even modest or small development can have large impact. Artefact patterning can only be fully understood through consistent recovery across a wide area, and the potential of a site may be significantly reduced through fragmentation even if substantial areas remain undeveloped. Hence in the Edgehill survey it proved impossible accurately to position all the battalions in the principal deployments, or fully to grasp important detail of the main infantry action, because the central area of the battlefield was so heavily disturbed and fragmented. With this said, Edgehill has shown that where fragmentation has taken place a site can still have a high archaeological potential and may yield terrain or battle evidence critical to the validation of hypotheses on the location and nature of principal deployments and the character and distribution of the action. Such analysis of fragmented patterning should become increasingly practicable as detailed research on well preserved battlefields enables us to distinguish the finer detail of particular aspects of the action.
With a 17th century battlefield a low density scatter can often reflect a particular character of action that did not involve an intense or indeed any kind of fire-fight. Yet when the small quantity of bullets or other artefacts is viewed within a wider context, it may reveal essential information about the nature of that action as compared to adjacent sectors. Thus at Edgehill the royalist right wing cavalry attack left a very low density of bullets, but also a very distinctive assemblage of calibres and the patterning, which when viewed with the adjacent areas suggested the position and orientation of the parliamentarian cavalry deployment and the direction of the royalist attack. The resulting re-interpretation placed the deployment in a different location to any previously suggested.
[…] large-scale topsoil disturbance will almost always distort patterning by the removal and redeposition of artefacts. As has been seen from Edgehill, distribution patterns that relate to firing lines and the firing of case from artillery can be highly specific. Thus the removal of topsoil over a 20m wide corridor could destroy the orientation of a case shot scatter. A good example of this is seen with the rail line on the Edgehill battlefield which has cut through the centre of the only such scatter so far identified on the parliamentarian left wing of cavalry. As a result, the orientation of the case scatter remains in some doubt; these data are critical to the exact alignment of the parliamentarian cavalry wing.
[…] in Grave Ground Coppice, a key surviving area in an otherwise destroyed zone at the heart of the battlefield, it proved impossible to undertake consistent survey on 10m transects because of the close spacing of trees and the density of undergrowth and roots.
Understanding case shot is likely to be central to the analysis of 17th century battlefields, because typically two artillery pieces were placed between each infantry battalion in the front line of the battle array which, at circa 150 m or closer, fired case shot when receiving an enemy attack. Thus, the spread of case shot can assist in the exact positioning not only of the battle array but also of each infantry battalion in the front line of an army following defensive tactics (e.g. Young 1967: 289), and accurately locating deployments is the essential starting point for any analysis of a 17th century battle (Norton 1628: 138-9). It may also, if patterns are not too heavily superimposed, clarify later phases of action. This is because, unlike all other bullets for which there can be no clear determination of point of origin, each case contained up to 100 projectiles, sometimes more depending on the piece being used, and firing resulted in a linear spread fanning out in the direction of fire.
In the case of Edgehill the full survey of the surviving and accessible battlefield, covering some 5 km², took nearly three years of part-time work comprising more than 3500 man hours of detecting at 10m spaced transects.
[As the] primary objective is [was] to systematically survey the whole of a large battlefield in a limited time and with modest resources […] Edgehill was surveyed at 10m spaced transects as this proved to be the lowest intensity which still enabled meaningful patterning to be recovered in the bullet distribution. With a detector sweep of 1.5–2.5m on each transect, depending on the technique of the detectorist, and with a forward prospecting speed averaging about 12m per minute the survey achieved near complete surface scanning of each transect, and so it can be argued that the 10m spaced transects represent a 15–20% sample of the surface area. After completion of the battlefield-wide survey more intensive sampling at 2.5m spaced transects was undertaken to refine the detail of the patterning in key areas, as for example with the case shot scatters.
In order to validate the bullet classification developed for the survey, assemblages from wreck sites […] were examined to provide comparative data of unfired bullets. […] This was complemented by experimental firing of both small arms and of artillery using case shot, to reproduce the firing damage to be expected.
The [experimental test] firing also enabled the final, archaeological range of bullets to be established, that is where the bullet finally ended up after missing its target. Thus for example bullets fired point blank at shoulder height on level ground on grass from a 12 bore musket towards a target standing at 90m distance would, if they missed, bounce at about 200m and bounce and roll again several times coming to rest at circa 300m. With case shot fired from a minion the width of spread of bullets was found to be circa 50m and the length of the spread was from within a few tens of metres from the gun to more than 300m from the gun.
Some cavalry carried carbines as well as pistols […] They [the carbines] are very sparsely associated with the wider cavalry action, concentrating particularly in association with musket bullets, both in the core infantry action and in the area of presumed attack on the baggage train in Little Kineton. This suggests they were being used as the more favoured weapon by cavalry of both sides when fighting against infantry. A significant revelation of the Edgehill survey is the sheer quantity of carbine and especially of pistol calibre bullets. This is in contrast to the evidence previously published for Naseby [a battle in 1645] […] Most such non-systematic investigation [establishing a consistent broad survey] may grossly under-represent the smaller calibres, giving a highly distorted picture of a battle.
127 bullets were recovered which had been fired as case shot from artillery. The sampling rate achieved with the 10m base survey means that several scatters were represented by just one or two bullets and thus it is possible that some locations where case was fired were not identified at all. The distributions were only clearly resolved where intensive resurvey at 2,5m interval was undertaken. This revealed the orientation of most scatters and suggested the approximate gun position, based on the experimental firing results. This case shot evidence proved critical in reconstructing the location and orientation of the original parliamentarian battle formation when it received the royalist attack, for the artillery were typically placed in pairs between each battalion. At Edgehill there were also three pieces with the left wing of cavalry, one of which appears to have been located in the survey.
[…] the work at Edgehill has provided a detailed example of the way in which the large scale application of the techniques of battlefield archaeology can transform the understanding of historic battlefields of the 17th century.
The distribution of pistol, carbine, and musket calibres appears broadly representative of the intensity and character of fire-fights, and though not all fighting involved the use of firearms, the patterning may be expected to reflect the location, extent, and intensity of most of the action. Musket calibre bullets are concentrated in the centre of the battlefield, but they represent only part of the area of infantry action, much of which has been destroyed by modern development. The scatter of musket calibre bullets also extends north-westwards at low density, a pattern interpreted as the rout of parliamentarian infantry.
The scatter increases again near Little Kineton, where it is mixed with a higher density of pistol bullets. This grouping is interpreted as the defeat of troops guarding the baggage train, positioned around Little Kineton, in the face of a royalist cavalry attack. The bandolier items also have a distinct distribution that appears to relate primarily to this area of infantry rout and defeat. A further scatter of musket bullets, towards the east of the battlefield, appears to represent the final stages of the infantry action, when the royalists retreated in the face of a parliamentarian counter-attack. According to the battle accounts, this was finally halted by artillery and musket fire from a ditch.
[…] revealing the orientation of most scatters [ofhailshot/ case shot] and, with the benefit of experimental firing results to aid interpretation, in several cases suggesting the approximate gun position. This hailshot evidence proved central in reconstructing the location and orientation of the original parliamentarian battle formation because, in this period, field artillery was typically placed in pairs between each battalion. At Edgehill, in addition, three artillery pieces were placed with the left wing of the cavalry. Only one off these three appears to have been located in the survey.
In the Edgehill survey all three elements of analysis have been employed: historic terrain reconstruction; re-analysis of the documentary accounts of the battle, enabling the deployments and events to be placed within the terrain reconstruction; and large-scale survey of the battle archaeology, leading to a revision of the overall interpretation. This work suggests that the battle lines need to be re-orientated by 90 degrees compared with those shown by studies based principally on the written accounts. It also appears to locate securely various elements of the action in relation to the terrain.
Despite the level of detail available from Edgehill, considerable uncertainty remains. While this is partly due to the degree of loss in the heart of the battlefield, it is also because of the current lack of comparative evidence. When other well-preserved Civil War battlefields have been explored in similar detail a more sophisticated interpretation of artefact distributions may be possible, which in turn may necessitate reinterpretation of Edgehill.
The Edgehill Survey (2004-7): Between 1st August 2004 and 23rd September 2007 a total of 144 days of metal detecting were undertaken, comprising 575 man days or over 3500 man hours, including setup and clearance time.
A base survey comprising transects spaced at 10m intervals fulfilled the first objective, of gaining a consistent picture across the whole battlefield to establish the extent and broad character of the action. This was followed by more intensive sampling in selected areas, to reveal detail of particular elements of the action. […] The base survey took 432 man days of fieldwork to cover 489 hectares.
[…] These issues [concerning artillery pieces, their numbers and relative positions within the battle arrays] could prove significant in the analysis of the battle, but any reassessment must await the results of the research by Marsh [Simon; in personal communication with the author] on both the numbers of troops present and the numbers and types of artillery piece. This work is being integrated as part of a reanalysis of the battle in significant revision of the numbers, formations and frontages presented here. When the numbers and type of artillery pieces with the army are better defined then this may also have implications for the interpretation of the hail-shot scatters discussed […]
In the absence of relevant documents it has proven impossible to identify with absolute certainty those enclosures and hedges already in existence in 1642.
Nineteenth-century re-planning, modern housing infill and the earthmoving undertaken to create the sports field [between Little & Great Kineton] will almost certainly have destroyed significant archaeology. [—…] massive destruction caused by construction of the Kineton munitions depot across the battlefield in 1943-4 and then it's rebuilding in the 1980s. The 1940s silos and the railways which served them caused total destruction in the areas directly affected and occasionally over small adjacent areas where additional earthmoving also took place […] The degree of survival of ridge and furrow on the 1947 air photos provides a vital record as to where earthmoving had and had not taken place, which is critical to determining where there has been destruction of the battle archaeology and also redeposition. […] in the 1980s […] a total of 24ha was restored to agriculture, involving substantial earthmoving. […] the disturbance did not generally extend beyond the areas destroyed. […] the battlefield remained almost wholly under pasture from enclosure in the eighteenth century through to at least the late 1940s.
In 2002 Glasgow University Archaeological Research Division [GUARD] conducted an intensive metal detecting survey of three small areas […, which] may have removed so many finds that all other distribution plans from subsequent survey work […] will be distorted, without sufficient data being available to enable a corrective calculation to be made. This is a particular problem in the area immediately adjacent to the depot, for here the GUARD grid runs along the north eastern boundary [approximatelyLAT/LONG 52.14052, -1.47850] of the highly important scatter of artillery hail-shot bullets.
However, as most but not all fighting involved the use of firearms, it [bullet archaeology] will not fully reflect the location, extent and intensity of action.
Musket calibre bullets are distributed across most of the battlefield in small numbers, though there are several very distinct gaps, particularly within the zones of cavalry action. Where they lie on the south-east side of the battlefield, associated with a few fragments of hedgerow, and in the enclosures in the northern edge of the battlefield they may represent the action of dragoons and musketeers deployed in hedgerows.
The more consistent and extensive spread of musket calibre bullets across the centre of the battlefield represents the infantry action. It can be divided into two halves, each apparently associated with a slightly higher density focus towards the centre. In the western half of the field the scatter has a distinct northern and eastern boundary but to the south it largely disappears into the unsurveyed and destroyed areas. On the eastern half of the field the scatter running east-north-east seems to have a very distinctive northern and southern boundary. Where of a lower density, running westwards and eastward, these scatters may accord with the flight of infantry (parliamentarian to the west and royalist to the east).
For the bullets above 22g that are certainly hail-shot there are three broad groups by mass […]. When the three groupings are separately mapped there is no clear separation of patterning. This might mean that scatters from different guns overlap or that hail-shot loads were of mixed calibre bullets. […] the scatter close to Little Kineton has not been resolved into a coherent, orientated pattern of firing. This is unfortunate as the scatter demonstrates artillery fire in an area where none is recorded in the primary accounts. This and the presence elsewhere of several other isolated hail-shot bullets raise questions about the wider use of hail-shot on the battlefield. These questions can only be resolved, if at all, by far more extensive 2.5m resurvey.
The distribution of hail-shot has proved critical to the interpretation of the battle, because so much of the area of infantry action has been lost. In the Civil War they were used at close quarters, had a relatively short final range and were fired from pieces which were typically positioned within or immediately in front of the infantry battle array. Most often there were two battalion guns, one placed on either flank of the battalion when in battle array. Most importantly, a single firing produces a linear scatter orientated in the axis of firing, thus providing important information on the orientation of fire from the battalions. However, at certain times enfiladed fire was also used, and so hail-shot may occasionally have been fired at an oblique angle to the battle line.
While heavy guns in fixed batteries could also fire hail-shot if attacked, the main use will have been by galloping guns attached to the battalions. These were intended to be moved during the action and thus, as a battle evolved, it is possible that new gun positions were taken. This means that there is a potential for overlapping scatters and also for one gun to fire hail-shot in several places on the field at different times. However, it is the initial clash that is likely in most battle to have produced the most coherent, dense and consistent pattern of hail-shot.
For Edgehill hail-shot, with a mass of 30-38g, a maximum range of circa 150 - 200m might be expected, but with the most effective range probably half that. This is important information when considering the likely position of the attacking force when the hail-shot was fired. However, the nineteenth-century experiments were concerned with killing poser and hence do not indicate the final range of overshot bullets. This is what battlefield survey normally retrieves. The final range and spread of bullets has therefore been established by undertaking experimental firing.
The orientation of the scatters from individual [artillery] guns, in so far as they can be distinguished, suggest firing in an east-north-easterly direction, for the fire might be expected to come principally form the defending force, who had to be on the west side. This suggests a wholly new alignment for the parliamentarian infantry deployment, running from north-west to south-east. This is almost parallel to and just north-east of the old Banbury road [Red road], which may have been used to bring the troops and particularly the artillery out from Kineton to their firing positions. This places the centre and right of the parliament foot on Bladon hill, which is where the accounts seem to indicate that they stood. Not one account suggest that these infantry ever gave ground, so the main action should be just forward of their deployment. […]
The fine detail of the bullet scatter, in so far as it can be resolved in this fragmented context, shows that hail-shot [fired from larger artillery guns] tends to concentrate where the densities of musket bullets are less. There seem to be several boundaries within the scatters between types of bullet or where the density of the bullets suddenly falls off. These boundaries broadly correspond to the orientation of the hail-shot scatters, which all seem to run at about 90 degrees to the western and eastern boundaries of both the musket and the hail-shot scatters. When the royalist battalions, reconstructed to scale from de Gomme's plan [drawn for Prince Rupert's diaries after the war], are superimposed on this area they seem to broadly correspond to this patterning in the scatters.
The [Edgehill] study has highlighted various strands of research that may improve our ability to interpret battle archaeology of the seventeenth century. […] Using the battle archaeology, the level of detail achieved in the reinterpretation of deployments and action at Edgehill has far exceeded the potential suggested in the 1995 study of Naseby battlefield.
The battle archaeology for Edgehill provides a completely new strand of evidence, which has never before been applied to the interpretation of the whole of any of England's Civil War battles. When these data are placed within our reconstruction of the historic terrain, it becomes possible to explore the evidence in primary accounts from a completely new perspective.