3.4.2 Artefact distribution case study: equestrian equipment
by Hanne Sheeran

The VASLE 'National dataset' contains over 650 horse-related artefacts. Approximately 200 have been identified as Anglo-Scandinavian on the basis of their decoration, but this figure is likely to be higher because many entries have too little detail to be assigned to a culture or style. The main categories are stirrup terminals, stirrup-strap mounts, and bridle or harness fittings.

Stirrups are thought to have been introduced to England by the Vikings. In Scandinavia, the use of metal stirrups became common in the 10th century and this trend spread to England in the 11th century (Graham-Campbell 1991, 81). There is no evidence for the use of metal stirrups in England before the 11th century; however, the word 'stirrup' originates from the Anglo-Saxon word for rope, which suggests the earliest forms were not made of metal and therefore have not survived in the archaeological record (Seaby and Woodfield 1980, 87). The introduction of the stirrup marked an important stage in the art of fighting from horseback, as it allowed the rider a more secure position from which he could wield his spear or battleaxe thereby gaining greater value from the weight and momentum of the horse (Seaby and Woodfield 1980, 104). Complete stirrups are rare – only four have been recorded by the PAS – but stirrup terminals and stirrup-strap mounts are relatively common finds.

Stirrup-strap mounts were placed at the junction of the metal stirrup and the stirrup leather. They are normally 40-60mm long and 30mm wide with a prominent flange projecting from the lower edge of the undecorated, reverse, side and usually two or more fixing holes for iron rivets (Owen 2001, 209). There are 340 stirrup-strap mounts on the PAS database, including over 100 decorated in the Ringerike or Urnes style. However, Williams (1998) has catalogued over 500 stirrup-strap mounts as a result of cooperation with detectorists. He has classified them into two main groups based on shape: Class A are triangular and Class B trapezoidal or subrectangular (Williams 1998, 2). Originally thought to be book or box mounts, they were subsequently identified as stirrup-strap mounts owing to their association with other equestrian equipment in Scandinavian burials (Williams 1998, 3). The mounts have been dated to the 11th century because of their association with stirrups, and also due to the presence of Ringerike- and Urnes-style ornament on the objects. Most of the stirrup-strap mounts of Class A and a few Class B examples show elements abstracted from or influenced by the Ringerike and Urnes styles (Williams 1998, 8). This further supports the notion that the equestrian equipment in general can be associated with Scandinavian activity in England.

Little has been published on the bridle and harness fittings. However, the VASLE dataset shows that many of these were also decorated in the Ringerike and Urnes styles and therefore they can be associated with the stirrup-related objects. Like the dress accessories, most of these artefacts were low-quality, mass-produced objects and much of the ornament is somewhat debased (Fuglesang 1980, 44-5). Graham-Campbell (1991, 88) has suggested that there must have existed prestige horse equipment that was consequently imitated.

Previous research

Even before the advent of metal-detecting, the corpus of artefacts relating to Scandinavian equestrian activity in England largely consisted of stray finds, generally from rivers (Seaby and Woodfield 1980, 102). The absence of such finds from excavated and documented contexts resulted in them receiving relatively little attention from archaeologists (Seaby and Woodfield 1980, 87). Generally, it was accepted that the Vikings were responsible for the increase in the use of horses for military purposes at this time. The earliest examples of spurs in England have been found in 9th-century Scandinavian burials in Norfolk and Cumbria (Hinton 2005, 153-4). There are also numerous references in the Anglo-Saxon Chronicle to the use of horses by the Great Army in the 9th century (Davis 1989, 142). However, there is no evidence that horses were used in combat at this time and it is more likely that the Danes used them to travel quickly but dismounted to fight (Graham-Campbell 1991, 81).

In 1980 Seaby and Woodfield surveyed the stirrups of late Anglo-Saxon England. At that stage, the stirrup finds were concentrated in three areas: the Thames and lower Severn valleys, East Anglia and Lincolnshire (Seaby and Woodfield 1980, 102). The fact that only a third came from within the Danelaw led the authors to conclude that they had not been introduced by the Scandinavian settlers of the 9th and 10th centuries as was previously thought. However, the stirrups had a clear Scandinavian connection. Many were believed to have been imported from or via Scandinavia, and others had decoration comparable to the Anglo-Scandinavian stone sculpture of Yorkshire and the north Midlands (Seaby and Woodfield 1980, 98, 101). The authors argued that their distribution in fact correlated with the movements of the Danish leaders – Sven, Olaf and Cnut – in the late 10th and early 11th century (Seaby and Woodfield 1980, 102). Rather than representing specific recorded raids, the survey concluded that the finds represented the first indication of stirrup-using horsemen in Saxon England in the 11th century (Seaby and Woodfield 1980, 102).

Graham-Campbell added to Seaby and Woodfield's research by showing that the introduction of stirrups to England in the 11th century was 'complemented by the appearance of a range of horse-trappings, not previously used in England, showing design influences from Scandinavia in the time of Cnut' (Graham-Campbell 1991, 88). He argued that England owed much to the Danes for innovations and improvements in horse-equipment in the 10th and 11th centuries. Finds of equestrian equipment from England can be compared to finds from Scandinavia, particularly Denmark where the increase in equestrian graves in the 10th century is evidence for the rising importance of the cavalry (Graham-Campbell 1991, 81). He has suggested that by the 11th century the cavalry were equally important and enjoyed a similar status in England. The status of the rider would have been reflected both by the quality of the horse and the riding equipment itself (Graham-Campbell 1991, 77-8). According to Graham-Campbell (1980, 87): 'The finest examples of the decorated English group [of stirrups] have a predominance of scrolled ornament, reflecting that known on contemporary swords and spurs, as well as on spears and battle-axes – evidence of an integrated "look" for the fashionable equestrian warrior of the early to mid-eleventh century'. Graham-Campbell also associated the equestrian equipment more directly with the reign of Cnut because of the use of the Ringerike style, which was at the height of its fashion during his reign in the early 11th century (Graham-Campbell 1991, 82).

The most significant piece of research on the metal-detected equestrian equipment to date is Williams' (1998) catalogue of stirrup-strap mounts, which defines and classifies the mounts (above) as well as providing information on their distribution. Based on the decoration and the datable contexts from which the excavated samples came, it was concluded that the mounts were in use for a relatively short period of time, 'perhaps from the first quarter of the 11th century at the very earliest, to around 1100 or not long after' (Williams 1998, 8). The mounts catalogued by Williams have a wide distribution within England with concentrations of finds in Lincolnshire, East Anglia and much of southern England, and a patchier scatter of findspots across the Midlands. The artefacts were absent in the far north and north-west and also in Cornwall (Williams 1998, 14).

There has been some debate as to what the Scandinavian style equestrian equipment actually represents. Hinton (2005, 157) has argued that the widespread use of equestrian equipment decorated with the Ringerike and Urnes style is not in itself evidence for a significant Scandinavian population in England in the 11th century. This is due to the fact that there is some debate over whether these styles evolved in Scandinavia and also the fact that fewer dress accessories are known in these styles (Hinton 2005, 157). Richards (2004, 153) has suggested that these artefacts may represent an elite cavalry group with Scandinavian pretensions associated with the reign of Cnut.


The horse-related artefacts are distributed fairly widely across southern England with the main concentrations in Norfolk, Suffolk, Kent, the Isle of Wight and Hampshire. The finds are absent from the north-east and north-west and also from Cornwall, although this is not surprising with low numbers of finds recorded in those regions in general. The biggest differences between the equestrian equipment and the overall VASLE distribution are in Yorkshire, where there are few equestrian artefacts and those that have been found are all from the south, and the counties of the southern Midlands where there is a surprisingly low concentration in Oxfordshire, Buckinghamshire, Bedfordshire, Hertfordshire, and Cambridge. However, the artefacts are relatively well represented in the west Midlands where they run in a broad belt from Gloucestershire to Lincolnshire. The distribution is quite similar to that of Ringerike- and Urnes-style dress accessories, although there are greater concentrations in Kent and the west Midlands.

The VASLE distribution is radically different to Woodfield and Seaby's stirrup distribution. The number of finds in East Anglia, Kent, Hampshire and the Isle of Wight has greatly increased. However, the region best represented by Seaby and Woodfield's finds in 1980 – London, Oxfordshire and Buckinghamshire – is surprisingly lacking in finds on the PAS.

There are also significant discrepancies with Williams' distribution of stirrup-strap mounts. Hampshire, the Isle of Wight and Kent are much more heavily represented by the VASLE data, as is the belt of equestrian equipment running from Gloucestershire to Lincolnshire, which is almost non-existent in Williams' map. Williams, however, has significant concentrations that are absent from VASLE, such as in Wiltshire, Dorset, Oxfordshire and Berkshire.


The comparison of the VASLE distribution to earlier distribution maps has significantly altered what was known. In 1980 Seaby and Woodfield were able to associate finds of stirrups with confidence with the recorded movements of Sven, Olaf and Cnut in the late 10th and early 11th centuries. The VASLE distribution shows that the area in which their activities were concentrated is actually poorly represented, despite the fact that metal-detecting is common in this region. Clearly both sets of data have significant biases. Seaby and Woodfield's survey was heavily reliant on river finds; the data for this project are reliant on metal-detecting and therefore perhaps it is not surprising that the results are so different. However, the differences in distribution also draw attention to the danger of correlating findspots with specific recorded historical events, especially when the corpus of known artefacts is small.

The differences between the distribution of Ringerike- and Urnes-related equestrian equipment with Williams' general distribution of stirrup-strap mounts is interesting. Not only are the majority of equestrian artefacts on VASLE stirrup-strap mounts but both sets of data rely on metal-detecting and are therefore more easily comparable. However, the sources of data were evidently different and it is possible that the differences relate to the relationships between individual Finds Liaison Officers and their local metal-detecting communities, as well as Williams' own contacts and connections. Whatever the reasons, the differences in these two distributions illustrate the limitations of the PAS and its incomplete nature.

What is the significance of the widespread distribution of this category of artefact? Graham-Campbell (1991, 82) has argued that the equestrian equipment highlights the links between England and southern Scandinavia during the reign of Cnut. A clue to what the artefacts represent may come from the other side of Cnut's kingdom – Denmark. As Graham-Campbell (1991, 81) noted, cavalry became increasingly important in Denmark in the 10th century. This is most clearly represented by the 'cavalry graves' identified by Randsborg (1980). Unlike the stray finds recorded by the PAS, these graves contain equestrian equipment in context and therefore may provide insight into the culture and circumstances that they represent. The graves include riding equipment as well as swords and lances and are often in chambers, sometimes accompanied by horses (Randsborg 1980, 127). Geographically, the burials occur in a belt around the Jelling province, corresponding with the borderlands of the early Danish state, and are concentrated in two regions – north Jutland and a belt from south-east Jutland across to the island of Lolland (Randsborg 1980, 127). Randsborg (1980, 129) has suggested that the formation of this state resulted in the creation of a number of social positions unknown to the earlier society and regards the graves as representing military duties in the new state (Randsborg 1980, 129). They represent royal vassals and, in the north, possibly a landless military elite bound to the king (Randsborg 1980, 132-3).

Although the context of the finds in England is very different, it is possible that they also represent a new social group or military elite bound to Cnut. The association between the equestrian equipment and the crown is logical. In the early medieval period the use of large horse was the preserve of the elite owing to the high cost of maintaining a specialised stud farm, necessary to produce such horses (Davis 1989, 141). Documentary sources also attest to royal controls over horse breeding and the increasing importance of horses for military purposes in the late 10th and 11th centuries (Hill 2004, 114). In 1023 Cnut himself formalised the number of horses required by each earl and each thegn (eight and four respectively) (Davis 1989, 142), indicating his reliance on the cavalry.

According to Rumble (1994, 7), Cnut's reign opened the way to increased trade and political contact with Scandinavia and gave new social opportunities to Scandinavians living in England. Perhaps some of these new social opportunities came in the form of membership of a military elite and those that were a part of this expressed their loyalty to Cnut and possibly their background through the use of Scandinavian-style equipment.

The distribution of the artefacts may confirm their association with Cnut's reign and his military. For instance, there is a cluster of finds in Hampshire that may be related to Cnut's use of Winchester as a royal centre. The Isle of Wight became increasingly important strategically over the 11th century, which may explain the finds there. In the early 11th century the Anglo-Saxon Chronicle records its use as a winter retreat for Danish raiders, citing particular misfortunes in the years 1001, 1006, 1009, 1013, 1022 and 1048 (Tomalin 2007, 15). In 1053 the island was seized by Earl Godwin and later King Harold held it with his fleet (Tomalin 2007, 15). The belt of finds across the Midlands may represent a frontier zone. Rumble (1994, 6) points out that although Cnut nominally ruled over all of England, the unified English kingdom had only been in continuous existence since 959 and regional loyalties were still strong. 'The old differences between Mercia and Wessex, and Northumbria and Southumbria, were perhaps as great as the divisions between the Danelaw and the English midlands' (Rumble 1994, 6). The finds of the Midlands may represent a borderland that required the presence of military power. If this was the case then it would also explain the absence of these artefacts in northern England. The absence of finds in the area north of London may represent the fact that this region was somewhat stable and required little military presence.


© Internet Archaeology/Author(s) URL:
Last updated: Tues Apr 21 2009