Eagle Hill Masthead

Journal of the North Altantic

    Aim and Scope
    Board of Editors
    Editorial Workflow
    Publication Charges

Co-published Journals
    Northeastern Naturalist
    Southeastern Naturalist
    Caribbean Naturalist
    Neotropical Naturalist
    Urban Naturalist
    Eastern Paleontologist
    Eastern Biologist

Eagle Hill Institute

About Journal of the North Atlantic


Antiquarians, Archaeologists, and Viking Fortifications
Ben Raffield

Journal of the North Atlantic, No. 20 (2013): 1–29

Full-text pdf: complete with cover or manuscript only (Accessible only to subscribers.To subscribe click here.)


Site by Bennett Web & Design Co.
Antiquarians, Archaeologists, and Viking Fortifications Ben Raffield* Abstract - This article addresses the depth of our knowledge regarding Viking fortifications in England, Scotland, and Wales, assessing perceptions of them as a monument type. This sudy includes the investigation of antiquarian influences upon the interpretation of these sites. It is suggested that archaeological knowledge of these monuments is largely fragmentary, and that in some cases, current understanding can in fact be based on interpretations dating back as far as the 17th or even 16th century. Additionally, it is proposed that Viking fortified sites do not exist with any form of homogeneity as a monument type. The research process of these investigations, findings, and two case studies are summarized. The article discusses the current state of knowledge regarding Viking fortifications and suggests how the study of them should proceed. *Department of Archaeology, University of Aberdeen, St. Mary's building, Elphinstone Road, Aberdeen, AB24 3UF, UK; B.Raffield@Abdn.ac.uk. Introduction: The Viking Age and the “Invasion” of Britain Throughout much of the period known as the “Viking Age”, Scandinavian groups arrived to raid and campaign on British soil. These forces required a safe place to overwinter or situate themselves while under threat from indigenous armies and populations. In England, they first “appear to have made use of natural islands, such as … Sheppey and Thanet” (Richards 2004:38) and continued to do so throughout the 9th, 10th, and 11th centuries. The 991 Battle of Maldon, Essex, supposedly involved the honorable but foolish act by Ealdorman Byrhtnoth of “[giving] too much room to those wretches” (Griffiths 1991:55) and allowing the opposing Viking army to cross from their base on Northey Island to engage in a “fair fight” on the Anglo-Saxon-held shore. Viking armies are also recorded as constructing purpose-built fortifications, with the locations of these occasionally being recorded with some accuracy. Despite this, there have been limited attempts to identify and excavate these sites, though in the past few decades some inroads have been made with the unexpected location of the Viking overwintering camp at Repton, Derbyshire, used during the period 873–874 (Biddle and Kjølbye-Biddle 1992, 2001). Current work being undertaken by the universities of York and Sheffield at Torksey, Lincolnshire, the location of the same army’s overwintering in the previous year, has identified evidence likely representative of the Viking occupation, though the majority of the discoveries is thus far confined to small finds recovered largely through metal-detecting activities (University of York 2012). The evidence recovered to date from Torksey (Blackburn 2002:92–93, 2011) and the finds excavated at Woodstown in Ireland (Russell et al. 2007) have revealed tentative evidence that occupation and fortification sites may be linked with local, regional, and national economies. At Woodstown, for example, 208 lead weights have been found, which suggests an intrinsic relationship between the site and the wider economy—the presence of hacksilver also presents a definitive Scandinavian signature, which is not widely seen in Irish contexts (Russell et al 2007:25). Similar finds have also been recovered at Torksey (Blackburn 2002), suggesting that Viking overwintering camps or fortifications may be significantly more enigmatic than previously perceived. Though the scale of Viking “invasion” and the size of the armies involved has been subject to debate in the past, most significantly by Sawyer (1971) and Brooks (1979), this is still a divided subject. While Clarke (1999:40) states that armies probably numbered “hundreds rather than thousands”, “it does seem hard to reconcile contemporary accounts of the largest Viking forces with numbers below the low thousands” (Williams 2008:195). By attempting to locate sites associated with conflict, such as fortifications, we can begin to accommodate the “violent” side of Viking Age life within our current views of Viking Age society, which have been augmented by development-instigated excavation at “domestic” sites such as Coppergate and Dublin. If we ignore the “military” aspects of Viking life, then we are in danger of creating a pacified past by which only the activities attributed to the more sociably agreeable side of Viking life are sufficiently understood. Indeed, it has already been argued that we have “lost sight of the violent side of the Vikings” (Price 1991:7), something that has only begun to change in the last few years. Our substantial knowledge of life in the Roman army, for example, is taken from extensive excavations at military installations and fortifications such as Inchtuthil Legionary Fortress and Housesteads Fort (Pitts and St. Joseph 1985; Rushworth 2009a, b). If these more ephemeral monuments dating from the Viking Age could be identified, then it may be possible to gain a clearer picture and greater understanding of how 2013 Journal of the North Atlantic No. 20:1–29 2 Journal of the North Atlantic No. 20 the peoples that built and occupied them lived and the impact that they were to have in the countries within which they were operating. A desktop study was therefore undertaken to assess the depth of our knowledge regarding Viking fortifications and to assess our perceptions of them as a monument type. The State of Knowledge Available literature on the subject, as has been alluded to, is relatively scant. Conflict in the Viking Age is broadly covered as part of general historical and archaeological texts (e.g., Forte et al. 2005, Hall 2007, Logan 2005, Richards 2004, Williams 2008, among others), though detailed case studies are understandably lacking. The discussion of fortifications within these studies is also therefore limited to references to sites recorded in The Anglo-Saxon Chronicle or those that have been tentatively identified in the field, as will be seen below. It is worth noting that there are no synthesized archaeological studies dealing with Viking Age fortifications across Britain at present, suggesting that there is a gap in our knowledge that requires filling if we are to understand how Viking armies operated on a day-to-day basis. Various authors discuss the subject of fortifications as part of a wider agenda (Abels 1997, Haslam 2005, Richards 2004, Williams 2008), though they are not addressed in any detail as an archaeological monument type. This depiction of the current state of knowledge should not be taken as a criticism of the above publications, instead being taken to merely highlight that these sites do deserve individual attention and investigation if they are ever to be really understood. Therefore, investigating our current understanding regarding these sites in terms of past antiquarian study is an ideal starting point. Though, to some degree, the influences of antiquarian study on our knowledge of the Vikings are recognized in academic sources (Lavelle 2010:229, Richards 2004:38), they have never been explicitly acknowledged or researched in past literature. In Ireland and on the European continent, however, Viking fortifications have received more extensive study and excavation. The longphort (pl. longphuirt [meaning literally “ship-base”]) sites in Ireland have been extensively investigated, with sites identified at Dunrally, County Laois (Kelly and Maas 1995, 1999) and Athlunkard near Limerick (Kelly and O’Donovan 1998), while a site at Woodstown, County Waterford has been subjected to excavation (Russell et al. 2007). Another site at Annagassan, County Louth is currently under study (Linn Duachaill 2012). Sheehan (2008) provides a useful overview of a number of longphort sites, while Maas (2008) considers them from a historical perspective. The results of the excavations at Woodstown have revealed a substantially defended D- or B-shaped enclosure on the bank of the River Suir. One of the defensive ditches, however, was allowed to rapidly silt up, having a smithing hearth constructed within it (Russell et al. 2007:32), while the site has also yielded evidence linked to textile production (Russell et al. 2007:37). Coupled with the large amount of lead weights recovered, these findings suggest that occupation may not have been exclusively “military”. The discussion of the material evidence in Ireland has run alongside debate regarding the nature of longphort sites themselves (Gibbons 2004). On the continent, fortifications in Brittany occupied during the Viking Age have also been subject to study, with excavations at Camp de Péran revealing ephemeral evidence for structures (Price 1991:10) as well evidence supporting hypotheses that the site was constructed and occupied or attacked by Vikings, possibly at some time in the 10th century (Nicolardot 1984, Nicolardot et al. 1987, Price 1991). At Trans, Ille et Vilaine, a site named Camp des Haies has revealed hastily constructed defensive ditches (Price 1991:10). In Scandinavia, a project investigating Viking Age fortifications has also been undertaken over a number of years in an attempt to quantify and discuss evidence from the period (Hedenstierna-Jonson et al., in press). The Trelleborg-style fortresses in Denmark and Sweden have long been subject to study (Nørlund 1948; Olsen 1977; Roesdahl 1977, 1986) as have the defended settlements at Birka, Sweden (Hedenstierna-Jonson 2006) and Hedeby, which now lies in northern Germany (see Jankuhn 1933, Knorr 1924, Kramer 1999, Schietzel 1981a, among many others; for a recent summary written in English of past research, see Hilberg 2008). These sites are of a different character and often of a different date compared to known and postulated British sites, but must be considered if we are to place the latter within their correct geographical, socio-economic, and political contexts. In Britain, the most extensively excavated Viking fortification is the 873–874 overwintering camp at Repton, Derbyshire, where a substantially defended D-shape enclosure was located on the south banks of an old course of the River Trent (Fig. 1). The defenses included a fortification ditch over 8 m wide and 4 m deep, the upcast of which would have formed the interior defensive bank. The Anglo-Saxon church of St. Wystan was also incorporated into these defences, presumably as a fortified gatehouse. The location of a number of Scandinavian burials exhibiting signs of violent trauma and a mass grave within a reused Anglo-Saxon mausoleum has been taken as suggesting that this was the site inhabited by the Viking Great Army during their overwintering 2013 B. Rafffield 3 in Repton (Biddle and Kjølbye-Biddle 1992, 2001). Despite being the best-understood Scandinavian fortification of the period in Britain, elements of the site remain perplexing. It is also postulated that the enclosure encountered may only be the center of a larger overwintering camp (Williams 2008:198), but this has not yet been identified. Anglo-Saxon fortifications, in contrast, are far better understood, with the fortified settlements, or burhs, constructed by Alfred of Wessex and his successors having been studied for some time; these continue to be investigated by both archaeologists and historians (Bassett 2007; Brooks 1964,; Haslam, in press [a], [b]; Hill and Rumble 1996; Kitchen 1984; Lavelle 2010; among many). The efforts of the Beyond the Burghal Hidage Project based at University College London have more recently aimed “to provide the first systematic study of Anglo-Saxon military organization and its landscape context for the period ca. 850–1066” (UCL 2005), the publication of which is eagerly anticipated. Additionally, sitespecific research at Goltho, Lincolnshire (Beresford 1987) has revealed a series of occupations dating from the Middle Saxon to Norman periods, including the construction and enclosing in substantial defenses of a manorial site in the late 9th or early 10th century, though the dating sequence at this site has been questioned (Reynolds 1999:130). Additional attempts have been made to understand Anglo-Saxon warfare on a larger scale (Lavelle 2010). Publications on the subject of Viking warfare in general are much more limited in number. Griffith’s (1995) attempt to study Viking warfare is largely problematic, with the author failing to grasp the context of his subject and enforcing his argument from a modern military viewpoint, despite making sound points at times. Similarly Siddorn (2005) attempts to study Viking military strategies and tactics based on his experience as a re-enactor, though he does stress that the book is not primarily designed to be academic (Siddorn 2005:7). Attempts to tie down the location of Viking Age battlefields have been largely unsuccessful, though a recent volume (Livingston 2011) focusing on the 937 Battle of Brunanburh has postulated a likely location in the Wirral. Viking Age warfare is therefore dominated by views related to popular misconceptions, stereotypes, and a general acceptance of these, though past work by the author has sought to reassess our knowledge of the subject and construct a framework for identifying potential sites of Viking Age conflict (Raffield 2009). Similarly to recent investigations into Anglo-Saxon warfare, wider-scale reanalyses of the subject matter are being currently undertaken by Williams (in press). It is important to recognize the significance of contemporary sources when attempting to study warfare during the late Saxon period and Viking Age. The year-by-year accounts in documents such as The Anglo-Saxon Chronicle, the Annals of Ulster, and continental sources such as the Annals of St. Figure 1. The fortified Viking overwintering camp at Repton, Derbyshire. Adapted from Hall (2007:85). 4 Journal of the North Atlantic No. 20 Bertin have done much to inform us of the events of the period, though the inevitable bias entwined within the passages must be acknowledged, as it was likely those writing these chronicles who were largely subject to initial Scandinavian attacks. To understand the fortifications constructed and used by various Viking armies or warbands is to better understand Viking Age warfare and the Vikings themselves. The excavations at Woodstown and work at Torksey have shown that fortifications and overwintering camps may be the locations of substantial commerce and not merely a site where an armed group chose to situate themselves for the winter months (Russell et al. 2007) or on a more permanent basis. The cessation of movement by an army would mean that the group had to be supplied, as living off the land would become increasingly difficult with time, initiating a series of new logistical difficulties that had to be dealt with peacefully, through force, or a mixture of both. Methodology The methodology undertaken was to attempt a survey of historic environment records (HERs), urban environmental databases (UADs), sites and monuments records (SMRs), and the national monuments record (NMR) in England, Wales, and Scotland. The search involved 82 records operating on a city, county, and regional basis in England in addition to four archaeological trusts in Wales and 16 SMRs in Scotland. Even from the early stages of the study, the approach proved time consuming and problematic—NMR records often included results dating before and after the Viking Age and others of dubious quality which had to be filtered out from the more suitable material. The search of HERs was undertaken at a relatively quicker pace; a standardized email was sent to every HER in conjunction with an online search of data undertaken where possible. For Scotland and Wales, the author was redirected to the online sources Canmore (http://canmore. rcahms.gov.uk/) and Coflein (http://www.cbhc.gov. uk/LO/ENG/Search+Records/Explore+Coflein/), respectively. Finally, the journal Medieval Archaeology was investigated for the years 1986–2007, and the Archaeological Data Service was searched. The latter provided a good example of the success rate encountered throughout the search—of 130 sites containing “Early Medieval” evidence, only one provided evidence of a fortification. This site was a possible Anglo-Saxon fortification in Athelney, Somerset, which has strong links to Alfred the Great and was investigated by Gaffney and Gater (2003). When constructing the methodology, the merits of a deductive approach were also considered. This analysis would involve the use of comparative examples from both inside and outside Britain and the identification of arguably “Viking’” features at such sites, combined with our knowledge of warfare in the period to form a “model” from which to investigate sites that corresponded to the criteria in order to assess the probability of their belonging to the Viking Age. Such an approach was not utilized for a number of reasons. Firstly, the construction of a model assumes that Viking fortification sites are a discrete and homogenous group of monuments that can be identified not only across geographical space but also the centuries spanning the Viking Age. Secondly, it also assumes that Viking armies or warbands would construct a uniform fortification. While Roman fortifications can be relatively easily identified across Europe due to their overtly military nature and construction adhering to a rigorous doctrine, we cannot expect the same of Viking armies, as they were susceptible to fluctuation in terms of size (Downham 2008:342). It must also be remembered that prior to the 11th century these groups were certainly not “national” armies and as such would not subscribe to any codified military doctrine. When we therefore consider that there may be no such thing as a “uniform” construction of fortifications, can we really expect an applicable and recognizable uniformity to exist over the wider European continent which can be applied to Britain? Thirdly, this approach would have necessitated the study of many sites that have no known connection to the Scandinavian presence in England. The Antiquarian Influence on Viking Studies Postulated fortification sites noted in the HERs were investigated. These have previous claims of association with the Scandinavians, either through modern scholarship, antiquarian studies, folklore claims, or associations with contemporary sources and annals noted in HER records or the aforementioned journals. This investigation was intended to first establish what we know about these sites and understand the influences that governed interpretation in the antiquarian past. There is no set date as which to ascribe “antiquarian” writing to—indeed, in England, a rising concern in the material evidence of the past began in the 15th century. It is not possible to summarize the development of antiquarianism and its entire impact on archaeology in these few pages, but excellent summaries are available elsewhere (see Chippindale 1994, Piggott 1989, Schnapp 1996, Trigger 2006, among others). A few points, however, must be noted here. In England especially, folklore, anthropology, and linguistic data heavily influenced archaeological interpretation (Trigger 2006:138); 2013 B. Rafffield 5 the Vikings held a prominent place within antiquarian studies from the 16th century, when “politics and religion inspired an interest in the good old days of an independent Anglo-Saxon state” (Hall 2007:218). Any understanding of prehistory was very limited, with scholars having “no sense of chronology apart from what could be ascertained from written records” (Trigger 2006:86). Prehistoric remains were arbitrarily ascribed to the Viking presence in England, with even Stonehenge at one point being postulated by some to be a Danish construction due to comparison with the “Hunnebedden” (long tombs with massive capstones in Holland) that were thought of as “prototype” versions of Stonehenge (Chippindale 1983:61). Following correspondence with Danish antiquarian Olaus Worm (1588–1655), Dr. Walter Chaleton (1619–1707) postulated that similar monuments in Denmark known as “dysser” were such prototypes for Stonehenge. Given that the Danes were known to have invaded Britain in the 9th century, it was proposed that the idea migrated with them (Chippindale 1983:61), resulting in the construction of Stonehenge. The Saxons and Romans, who were also well-documented invaders and settlers of Britain, or the ancient Britons whom the Romans had encountered, were also regularly evoked when attempting to associate monuments with certain periods (Trigger 2006:86). This paper discusses 40 sites that HER records show to have been postulated to have been used by the Vikings during their time in Britain (see Fig. 2 for map, Appendix 1 for gazetteer), the associated sources suggesting that archaeological remains related to 25 of these had been first identified as Viking by antiquarians. Certain regions of Britain were much more heavily represented than others, with the east of England especially yielding a large number of sites (17 from Bedfordshire and Essex alone). Scotland yielded only a single site from the Outer Hebrides, while two sites were located in Wales: one of which is associated with the 893 battle at Buttington, Powys, and another with a fortified settlement at Llanbedrgoch, Anglesey, which may have been under Viking occupation (Redknap 2008:406). As Appendix 1 will show, however, not all sites that have been identified as Viking fortifications by antiquarians have been followed up by modern scholarship, the limitations and speculative nature of the available sources being recognized. In other cases, antiquarian writers may indeed have observed features that represented the extant remains of Viking camps or fortifications, but these have since been destroyed with no traces surviving in the modern day (see, for example, Spurrell [1885] on Benfleet and Boyd-Dawkins [1873] on Buttington). Whether these features truly did represent fortifications may never be known. Additionally, other sites such as the “Aldewerke” in Cambridgeshire have been identified in more recent times, and it is these that are often stronger contenders for being of early medieval date. In a few cases, however, possibly spurious sites identified in the antiquarian past have permeated the modern literature. While investigating the antiquarian influence on Viking studies and the sites in this paper, a chain of inter-relying interpretations was encountered with regards to an earthwork at Barton-Upon-Humber, Lincolnshire. This site was thought to possibly represent a Viking fortification or “burh” (Bryant 1994:73, Heritage Gateway 2006, Reynolds 2003:117), though Rodwell and Atkins (2011) postulate that it could date to the middle Saxon period. With little dating evidence available, the attribution of this site to the Vikings was based on place-name analysis, stratigraphic relationships, and comparison, with Bryant (1994:75) citing two comparative examples—Howbury (also known as Renhold) in Bedfordshire and Ringmere in Norfolk. Bryant references the work of Dyer (1972), who notes both of these sites as being of Viking provenance in an article regarding the “earthworks of the Danelaw frontier”, which looked to identify Viking fortifications mainly in the Bedfordshire landscape. Richards (2004:39–40) also cites these two sites, among others, suggesting that Dyer’s work has certainly had some influence on our interpretations of fortifications today. Given the context within which Dyer was working, however, his attempt to associate previously undated earthworks with the historically well-documented Viking presence in the Bedfordshire region would always be a difficult task—there are simply too few sources of reliable information on the subject and too few sites that have been subject to excavation. As such, he references and seems to have been heavily influenced by the 1904 Victoria County History for Bedfordshire (Goddard 1904), which features many of the sites mentioned in his article. Goddard, in turn, seems to have obtained information from older sources, personal correspondence, and at least in part from the works of John Leland (1503–1552). Writing on Renhold, for example, he notes that Leland stated that a number of skeletons were found between here and Bedford, which may have been Vikings operating out of the “Danish outposts” of Renhold and Willington (Goddard 1904:285). It is significant that as early as 1900, however, Armitage (1900:260) had recognized the dangers of relying simply on the interpretations of those past, stating that “it seems strange that in the nineteenth century any archaeologist of reputation should still follow the method of the archaeologists of a hundred or two hundred years ago, who first guessed at things, and then said they were so.” 6 Journal of the North Atlantic No. 20 Figure 2. Distribution of postulated Viking fortification sites identified through search of HERs. 2013 B. Rafffield 7 This example shows that through a reliance on older sources and a lack of targeted investigation it is possible that in some cases archaeologists have been inadvertently using 16th-century antiquarian interpretations as a basis for hypotheses, due to a reliance on these when county histories were being compiled in the early 20th century. When laying out the sites utilized in the study in a gazetteer (see Appendix 1), the antiquarian influence in some areas of the country was clear to see. Despite the heavy antiquarian influence in certain areas of the country, these interpretations should not be seen as an obstacle to be overcome by archaeologists but instead as a useful tool, though they must be considered with care. It is also likely that further antiquarian sources were used but not officially referenced in the older texts, making the identification of sources difficult. Given the chain of interpretations in Bedfordshire stretching back from Dyer (1972) to Leland, that Dyer’s “fortifications” continue to be referenced can only be problematic. Dyer’s bibliography also includes other early 20thcentury works such as Allcroft’s (1908) Earthwork of England and further illuminates his reliance on outdated sources, which has seriously impacted our interpretation of these sites in light of the absolute lack of excavation work that has occurred at many of them. Indeed, where excavation has taken place, sites such as Willington Docks (Fig. 3) have been reinterpreted as later medieval sites (Edmondson and Mudd 2004, Hassall 1975). The use of antiquarian sources should not, however, automatically lead to the dismissal of theories based upon such work. Such studies were largely based on field observations, folklore legends, and tales related to them by locals of an area and which often included a Viking Age influence. “Battle Hills”, Essex, for example, is believed to be a Danish burial ground due to the red-berried Danewort plants which grew there. Camden recorded in his (1607) work Britannia that the local inhabitants “still call [it] by no other name than Danes-bloud … [due to] the number of Danes that were there slaine, verily beleeving that it blometh from their bloud” (Sutton 2004). As such, antiquarian interpretations tend towards the rationalizing—making sense of local legends or monuments of unknown dates through the tales that were related to them. With regards to the relationship between archaeology and folklore, “folklore cannot be accepted on face value as portraying factual truths about the past. But neither can it be rejected as false … Acknowledging the historical dimensions of items of folklore can allow us to develop analytical approaches to their use as historical sources” (Gazin-Schwartz and Holtorf 1997:14). Figure 3. The “Danish Docks” at Willington, Bedfordshire. Excavation has shown this to be a later medieval site. Adapted from Allcroft (1908:386). 8 Journal of the North Atlantic No. 20 The approaches of archaeology and folklore in the past were united through the works of antiquarians and it is only with the rise of archaeology as a modern- day scientific discipline that we see interpretations moving away from the use of the latter, though it must be remembered that folklore remains important as part of an interdisciplinary study. Indeed, in Camden’s discussion of Bedfordshire, he makes note that Viking fortifications were constructed at “Temesford” (Tempsford), Benfleet and Shoeburyness in 917 and 893, respectively (Sutton 2004). This reference demonstrates that Camden possessed a knowledge of The Anglo-Saxon Chronicle and as such historical enquiry was entwined with that of the anthropological. This example also suggests that scholars may have possessed some kind of pre-disposed idea of what they were specifically looking for. A better understanding of these sites can only be obtained through focused research and investigation. For example, Risinghoe, Bedfordshire, theorized by Goddard (1904:296) to be a Viking observation platform or burial mound, has since been reinterpreted by English Heritage as a motte and bailey castle. Since we cannot be sure as to what form Viking fortifications took (with only one having been confidently identified in England), more intrusive investigation is likely necessary to try and establish the true nature of these sites. Though many years of experience go into reinterpretations based on field visits, the construction of a research framework within which to investigate a selection of sites may do much for our knowledge of early medieval earthworks. Despite the lack of study that had taken place, it seems that there are prevailing characteristics of a Viking site that seem to have emerged as being apparent to antiquarian investigators. These are attested to by Spurrell (1890) and Goddard (1904). Spurrell (1890:79) states that the earthworks forming part of a large enclosure at Shoeburyness, Essex, are “in accordance with the general mode of fortification [used] at that time both by Danes and Saxons” but does not elaborate on this. Armitage (1900:262) notes, however, that she is “not aware that any serious attempt has ever yet been made to ascertain what the nature of an Anglo-Saxon fortification was”, despite the fact that Spurrell evidently possessed some ideas regarding this. Shoeburyness is the location of a fortification constructed in 893 following the destruction of Hastein’s fortification at Benfleet. Though the substantially sized enclosure identified as the site by Spurrell has since been reinterpreted and is listed as a slight univallate hillfort (English Heritage 2011), it is possible that this was re-occupied by the Viking force. Unfortunately, much of the enclosure had been destroyed by a garrison by 1890, with only two stretches of earthwork still visible by 1903 (Fig. 4). The enclosure has also been heavily truncated by the sea. Goddard (1904:280) also highlights cultural indicators of Viking fortifications when describing the sites at Etonbury, stating that “the small mounds at the end of the ramparts are found in works reputedly Danish”, while Willington Docks has “certain unusual features, which appear to mark it also as Danish” (Goddard 1904:282). Laver (1930) also notes that at Pandal Wood, Essex mounds are incorporated into the earthwork banks (Fig. 5). It does seem, therefore, that rules of some form were constructed by various antiquarians by which they designated sites as “Viking”, though these are largely unknown today. As such, the modern interpretations based upon these may well be in serious need of a review, given thus far that only one site (Repton) can actually be considered to be of Viking origin and this Figure 4. The heavily truncated remains of the postulated Viking camp at Shoeburyness, Essex. Adapted from Chalkey Gould (1903:287). Plan courtesy of the Victoria County History, University of London. 2013 B. Rafffield 9 ly Viking (perhaps from comparison with sites such as Hedeby), and indeed this form has been confirmed archaeologically at Woodstown and Repton, though these two enclosures are of a very different scale. Many circular sites such as Renhold, Bedfordshire have been re-designated as “ringworks” by English Heritage, and Dyer (1972:231) admits that these are problematic: “there is a greater chance of confusing these sites with Norman ring-works, and indeed only excavation is likely to establish their identity with certainty”. Of the 40 sites in the gazetteer (Appendix 1), nine have finds associated with them which have been used to support theories of a Viking or Scandinavian connection. Some of these were recovered by antiquarians themselves and have contributed to their interpretations. At Gainsborough, Lincolnshire, for example, Allen (1834) noted the recovery of a “battle axe” among other finds, while at Frogmore Lodge, Hertfordshire, “swords” and “axes” were recovered (Pollard 1906). At Buttington, Powys, Boyd-Dawkins (1873) notes skeletons recovered from the churchyard as displaying battle trauma, relating these to the 893 siege and battle recorded in The Anglo-Saxon Chronicle (Swanton 2000:87). Finally, at Benfleet, Essex, Laver (1903:236) relates a local tale dating from the 19th century telling of burnt ships and human remains being found in the creek thought to date from the 893 battle in the vicinity. For other postulated sites, other stray finds are better recorded. At Blunham, Bedfordshire, an Anglo- Saxon spearhead was recovered in the vicinity of the site close to TL151 525 (Beds HER 9772), while more recently an Anglo- Saxon strap-end was recovered just to the east of the site (M. Edgeworth, University of Leicester, UK, 2011 pers comm.). Though the Viking fortification of 893 at Benfleet is as yet unlocated, a Byzantine coin dating 850– 950 (Essex HER No. 46854) was found in Benfleet Creek. Such coins are thought to have been brought to Britain by individuals site was only located by chance through modern archaeological investigations. Some supposedly “Viking” features are speculated upon, however. Cohen (1965:42) spuriously takes the existence of “inner and outer wards” at Willington Docks, Bedfordshire to be similar to those seen at Trelleborg. Goddard (1904:284) describes sheltered “boat nausts” at the site as typical Viking features—the “Northmen were accustomed to provide some such shelter for their fleets when campaigning.” Dyer (1972:229) asserts that the presence of D-shaped enclosures at the site bears similarities to the semi-legendary and as yet unidentified fortress of Jomsborg, which supposedly possessed separate fortifications for both the garrison and for shipping, though he notes that “it would be unwise to carry the similarity farther.” The importance that Dyer places on “harbour” sites is reflected by his identification of another in Bedfordshire at Clapham on the River Great Ouse, though this has since been lost to quarrying. As has been mentioned above, however, excavation at Willington has suggested later medieval occupation (Edmondson and Mudd 2004, Hassall 1975). It seems that the most important factor in recognizing a “Viking” fortification was the form of the site itself—a circular form being taken as representative of the Trelleborg fortresses, while a D-shaped fortification on a river is taken to be characteristical- Figure 5. An earthwork enclosure at Pandal Wood, Southminster, Essex, postulated to be of Viking construction. Adapted from Laver (1930:256) by kind permission of the Essex Archaeological Society. 10 Journal of the North Atlantic No. 20 Dyer (1972:226) records it as an irregular D-shape. From this study, it can be suggested that both Goddard and Dyer certainly possessed pre-conceptions as to the form that a “Viking” site should take. Goddard (1904) does not specifically highlight a D-shape as a Viking “signifier”, but it could be that this was seized upon by Dyer as some sort of indicative feature. His preconceptions may to some extent be based on the form of Viking Age enclosures in Scandinavia such as Hedeby, and he looked to locate sites conforming with this; indeed, Reynolds (2003:117) states that “the earthwork enclosures first considered by James Dyer … share, very broadly, a D-shaped plan form”, suggesting that Dyer was one of the first to identify sites based on their D-shaped form. Dyer also draws heavily upon the Trelleborg fortresses as a comparison for the Bedfordshire earthworks, though this argument is also less than convincing considering their relatively late date and the vastly different socio-political environment which was prevailing in late 10th-century Denmark when compared to late 9thand early 10th-century Bedfordshire. (Though it must remembered elsewhere that the Trelleborg fortresses pre-date the later phases of Viking activity in Britain. That Dyer’s article is titled “The Earthworks of the Danelaw Frontier”, however, surely places his focus in the late 9th/ early 10th century). Though circular earthworks with “a timber-faced earth rampart surrounded by a wide ditch … [have] excited interest” (Hall 2007:76) in the Low Countries, these are just as likely to have been constructed as defensive structures by the indigenous populations as by the Vikings themselves. These constructions nevertheless shows circular fortified earthworks to be possibly late Saxon or Viking Age, though the degree to which this can be considered to be diagnostic is questionable. In addition to shape, there also seems to be a concern with the area covered by the Bedfordshire fortifications. Of the seven sites recorded originally as part of Goddard’s (1904) Victoria County History, five of these are relatively similar in size—the diameter of the two circular sites he mentions differing only by 20 m. Three of the D-shaped earthworks are similar in size, with the site at Etonbury being the only one that is significantly larger, though the truncated nature of this site makes estimating its size difficult. The only significantly smaller site is Gannocks Castle, Tempsford (Fig. 7), measuring only ca. 60 m (200 ft) across. In the case of Gannocks Castle, the site was identified as a Viking fortification in an attempt to locate the historically recorded 917 fortification and battle at Tempsford. The small size of the site is justified as representing this historically important fortification, siege, and battle through the construction of a fictitious scenario by Goddard (1904:282). This interpretation involves the combined Huntingdon within Viking armies or via the long-range trading routes that the Scandinavians are known to have operated . Other features have been located through excavation, such as the furnished Scandinavian burials located at Repton, Derbyshire (Biddle and Kjølbye- Biddle 1992), while at Llanbedrgoch, Anglesey, hacksilver was recovered in addition to possible execution victims in the ditch enclosing the settlement (Redknap 2000). Finally, at The Udal, North Uist, the Scandinavian occupation is supported by pottery finds (Lane 2007), though the true nature and shortlived occupation of the small “fort” there is still not fully understood, though it “soon went out of use and was downgraded to a cabbage patch enclosure” (Graham-Campbell and Batey 1998:173). In terms of supporting a Viking presence, those finds that can be argued to be reliable indicators are few in number. The finds at The Udal, Repton, and Llanbedrgoch have been obtained through modern archaeological excavation, while the stray finds at Blunham and Benfleet, while convenient, cannot be taken as hard evidence of the hypothesized Viking fortifications there, despite nearby fortifications being attested in the historical sources, as they were not found in secure archaeological contexts relating to the sites. That such finds are being recovered, however, could be suggestive of a possible Scandinavian or late Saxon occupation in the vicinity. Regarding earlier finds such as those at Gainsborough and Frogmore Lodge, these must furthermore be considered in the context of what investigators were specifically looking to find—that is to associate these locations with a postulated Viking occupation. Case Studies: Bedfordshire and Essex As noted above, the study identified two concentrations of postulated “Viking” fortifications identified by antiquarians and later scholars, one of which unsurprisingly lies in Bedfordshire, while the other lay in Essex. The features of these sites were compared to investigate whether their interpretation could lie in the identification by antiquarians of certain “diagnostic” features that provided the reason for their being assigned to the Viking Age. In Bedfordshire, eleven possible sites were located by the study (Fig. 6), one of which has been hypothesized recently by Edgeworth (2008) to be the site of the 917 fortification and Battle of Tempsford. Of the remaining ten Bedfordshire sites, eight have been identified as having their roots in antiquarian interpretation. Five of these involved possible D-shaped earthworks, while four consisted of possible circular earthworks. The site at Manor Farm, Bolnhurst, has been significantly truncated to the point where it could have possessed either, though 2013 B. Rafffield 11 these earthworks as Viking. Goddard states that the army was largely destroyed at the walls of Bedford (a battle and Viking defeat being recorded in The Anglo-Saxon Chronicle in 917 [Swanton 2000:101]). and East Anglian army being previously encamped at Renhold (which possesses fine views west over the Great Ouse valley towards Bedford) and Willington, thereby also justifying the interpretation of Figure 6. The theorized Viking fortification sites in Bedfordshire obtained from the sear ch of HERs. 12 Journal of the North Atlantic No. 20 the north and west sides by the extent of Benfleet’s churchyard. Spurrell (1890) identified a large enclosure at Shoeburyness, postulating this to be the site of the 893 fortification (though, as mentioned above, this site is now considered to be a hillfort). Though this site has been heavily truncated by the sea, it would have been of substantial size, which would perhaps be expected of a fortification designed to house more than one Viking army. It is worth noting that reoccupation of an extant Iron Age earthwork by the Viking forces is a distinct possibility. A fortification identified in Pandal Wood, Southminster by Laver (1930) measures 116 m by 98 m, the form of the site being roughly pentagonal (Fig. 5). This site includes mounds built into the banks as well as within the interior (Laver 1930:257), which is a feature shared by Edgeworth’s hypothesized location for the 917 Tempsford fortification at Blunham as well as at the postulated longphort at Dunrally in Ireland. Laver attributes this site to the invasions of Knútr, though no finds were recovered at the site. At Danbury Camp, Chelmsford, first identified by Morant in 1786, Morris and Buckley (1978) determined that if a Viking occupation did exist at the 140-m by 50-m site, then this was largely aceramic. The site has been heavily truncated by market gardening, and as such, its plan is unclear. The final site is “Canute’s camp” at Canewdon, Rochford, recorded by Chalkley Gould (1903) as being “oblong” in plan and enclosing ca. 6 acres, though no upstanding remains of this earthwork survive. The site is visible on old maps, however, and can be seen attached to the Pastscape record for this site (English Heritage 2007). In contrast with the Bedfordshire sites, those in Essex seem to share a more rectilinear-based plan, though Rodwell and Atkins (2011:842) postulate that roads and trackways at Benfleet have fossilized a D-shaped circuit of defences. The irregularities are more pronounced than those that exist in the Bedfordshire sites. The size of the Essex sites is clearly more varied than those identified by Goddard in Bedfordshire. Two of the Essex sites (Pandal Wood and Canewdon) are arbitrarily ascribed to the 11th-century raiding armies of Knútr (Chalkley Gould 1903, Laver 1930), by which time the Danes The surviving remnants of this force are narrated as having retreated to Tempsford to construct a very small fortification to reflect the diminished size of the army. This site is conveniently identified at Gannocks Castle. In his description of Willington Docks, Goddard (1904:282) also notes that the “harbour” here would have been capable of housing “between twenty-five to thirty ships of the Gokstad type, which would allow for a force of about 2500 men”—a size that might be expected of two combined regional raiding armies and supporting the numbers postulated at the beginning of this article for larger armies. This example demonstrates the weight of historical sources in influencing perceptions of the Viking Age in the past; in some cases, there was a concern with interpreting the size and form of a site in order to associate a historically attested Viking army—that which was destroyed at the Tempsford fortification in 917—with otherwise unknown and un-interpreted earthworks of the size befitting an initially large and thereafter diminished force. Another cluster of five sites (the fortification on Mersea Island is as yet unlocated) was identified in Essex (Fig. 8), which was studied to provide comparative evidence to the Bedfordshire sites. With both counties being on the periphery of the “Danelaw”, it could be anticipated that the sites may share similar characteristics if indeed they were of Viking construction. The “fortifications” in Essex in fact vary greatly in size and form. Earthworks supporting Spurrell’s (1885) hypothesis for the location of the 893 Benfleet fortification are no longer extant, though he hypothesizes that the site was in part bounded on Figure 7. The medieval manorial site at Gannocks Castle, Tempsford, once hypothesized to be the 917 Tempsford fortification. Image adapted from Allcroft (1908:385). 2013 B. Rafffield 13 greater irregularity of the Essex sites is the fact that they are identified by four different antiquarians or archaeologists, which directly contrasts with the “Viking” sites in Bedfordshire, identified largely through the work of Goddard (1904) and later by had certainly been constructing fortifications to a circular plan in Scandinavia as seen with the Trelleborg fortifications, so the concern with these sites was probably not due to their being of a representative shape or form. What may be a key factor in the Figure 8. The theorized Viking fortification sites in Essex obtained from the search of H ERs. 14 Journal of the North Atlantic No. 20 Dyer (1972). The inland nature of three of the five Essex sites may also partially account for the greater irregularity in form, as the majority of Bedfordshire sites are associated with shipborne Viking forces. Both Bedfordshire and Essex feature sites incorporating mounds within “defences”, which has been noted at longphort sites in Ireland, as mentioned above. Whether this can be truly considered a “Viking” trait, however, cannot be confirmed as of yet. The varied nature of the postulated sites and the lack of excavated examples means that at present we cannot be sure as to what form a Viking Age, Scandinavian fortification in Britain would take. The conflicting evidence of Repton and Torksey seems to suggest, on one hand, that a Viking fortified site should be a heavily defended and overt symbol of military occupation, while at the same time, the occupation of a Viking army may in fact involve intensive trade, manufacturing, and the occupation of an area utilizing natural defensive features or no fortifications at all. That these two sites were used only a year apart indicates that there are unknown factors at play, which work at Torksey may be able to shed light upon. It could be significant that The Anglo-Saxon Chronicle states that the Viking army “took winter-quarters … [and] the Mercians made peace with the raiding-army” while at Torksey. The next year, at Repton, the army “took winter-quarters there, and drove King Burhred across the sea” (Swanton 2000:72). The presence and form of fortifications may have been directly related to military threat and the nature of political allegiances at any given time, further highlighting the highly fluid nature of the relationships between Viking armies and indigenous populations. The case studies in Bedfordshire and Essex show that certain features of sites may have been targeted as being perceived as indicative of a Viking occupation or construction. D-shaped enclosures feature regularly, especially in Bedfordshire, while the use of ships for which the Vikings are so famously known means that these sites are often associated with rivers or other bodies of water. As is noted with regards to Gannocks Castle and Willington, stories can be interwoven with these sites to explain why they take a certain form in order to correspond with historically recorded events. At this point, therefore, while the nature and date of these sites is so open to debate and interpretations are liable at times to be clouded by past speculation, constructing a prospective “model” of what a Viking fortification may look like from the sites included in the gazetteer (Appendix 1) is impossible. The discussion regarding the size and form of Viking fortified sites at present may run in a full circle, as despite D-shaped and circular fortifications having been archaeologically located both in Ireland and England and on the continent, there is nothing thus far to insinuate that these are decidedly Viking traits; indeed “D-shaped enclosures are found in pre-Viking contexts in England and in areas without the Danelaw” (Reynolds 2003:119). When all 40 sites across Britain were compared, D-shaped sites were a minority when compared to circular sites, and both of these occurred less frequently than sites of an altogether “other” form. Given that confidently identified sites in Ireland and Scandinavia such as Woodstown and Annagassan and Hedeby feature such enclosures, this is further evidence that a deductive methodology would not have been suitable for this particular study. The evident diversity of sites perceived to be of Viking origin should therefore suggest that archaeologists possess an open mind about the relationships between the physical appearance of a site and its origins (indeed, it must be considered that a substantial static army may not have used fortifications at all). The Bedfordshire sites also proved to be unique when compared to sites from other counties in that 7 of 11 sites were associated with waterways, suggesting again an evident concern in Britain to associate these sites with a certain idea of the way in which Viking armies were moving through the landscape. The raiding army which constructed the Tempsford fortification is recorded in part as moving south from Huntington (Swanton 2000:101); the quickest route to do so being by river, so could past scholars have possessed a concern with “identifying” sites specifically to accommodate theoretical Viking shipping? Willington Docks, for example, was perceived to have nausts to house up to 35 ships (Goddard 1904), while at Tempsford, Dyer (1972:225) postulates that “the Danes are more likely to have built on the west bank, with their boats on the river behind them”. As such, it seems that this particular Viking army was considered to be a shipborne force. The high number of riverine sites, especially in Bedfordshire, is likely reflective of this, though such sites were not confined to the county. The site at South Cove, Suffolk (Fig. 9) is situated adjacent to a former riverbed and thought to have included a quay (Morley 1924:173– 174). Morley (1924:174) goes on to compare this site to Warham Camp, Norfolk and also a Danish camp in Bedfordshire on the River Ouse. Therefore, while postulating that the site represents the Domesday sea-port of Frostenden, Morley is attributing a Viking construction to the site, the resemblance to other “Viking” sites and the location of the site by a waterway being the contributing factors to this interpretation. He spuriously dates the site to 876 and associates it with a Viking retreat following a defeat at Bloodmoor Hill (Morley 1924:177). Similarly, the sites at Canewdon and Pandal Wood, Essex, are both associated with the move2013 B. Rafffield 15 ments of Knútr’s army. In 1016, Knútr is recorded as “[turning] back up into Essex”, having been previously driven towards Sheppey by Ealdorman Eadric (Swanton 2000:151). It was after this that the battle was fought at Assandun (Ashingdon, Essex), which lies just under two and a half miles to the west of Canewdon. Again, here we may be observing the identification of a camp or fortification site based on its proximity with a historically recorded battle. Regarding the size of fortifications, it is also necessary to assume that, in addition to troops, the armies would have possessed animals, supplies, and other material associated with warfare. It is also likely that they were accompanied by associated camp followers. As such, the size of suspected sites must take this into account. For the fortification at Repton, for example, it is possible that various parts of the Great Army were billeted in the landscape around the overwintering camp that is represented by the D-shaped earthwork, or indeed that this earthwork may be only the central feature of a much larger, as yet unidentified enclosure. In light of this, the large site postulated by Edgeworth (2008) (Fig. 10) to be the 917 Tempsford fortification at Blunham may not be of unreasonable dimensions if the enclosure was to house the raiding armies of Huntingdon and East Anglia in addition to their combined stores and followers. The size of the “great army” and Viking armies in general has been considerably debated (Abels 2003, Brooks 1979, McLeod 2006, Roesdahl 1998, Sawyer 1971, among others). It is likely, however, that numbers fluctuated greatly depending on the number of groups partaking in a campaign, Figure 10. Site postulated by Edgeworth to be the 917 Viking fortification of Tempsford. The site is in fact situated just inside the adjoining parish of Blunham. The possible D-shaped enclosure within the red box (marked as “the castle” on old maps (Edgeworth 2008:8) measures some 200 x 80 m. Photo courtesy of The Heritage and Environment Service, Bedfordshire County Council, RAF 106G\UK\635.10\AUG 45\F20\1154\430. Figure 9. The Domesday sea-port of Frostenden, hypothesized to originally be a Viking fort and quay. Reproduced by kind permission of the Suffolk Institute of Archaeology and History. No scale available. 16 Journal of the North Atlantic No. 20 their status, and the rate of attrition that was suffered, as well as the regularity of which groups joined the larger force. The Anglo-Saxon Chronicle, for example, records that the “great army” divided in 875 and 877 (Swanton 2000:74), while Downham (2008:342) postulates that the leader Guthrum joined the army in 871. As such, we cannot expect fortifications to maintain uniformity in size even through the events of a single campaign, and it is likely that their size would fluctuate on a fairly regular basis depending on the size of the area chosen to be enclosed or whether any area was chosen to be enclosed at all. While fortifications may indeed take the form of D-shaped enclosures (this is an obvious choice for any group possessing ships wishing to safely enclose themselves), at present it is not possible to make judgements on a “Viking” site based purely on form and size. Even where similarities in form are shared, the vast difference in size of the Woodstown site and the known extent of Repton suggests that individual factors certainly come into play. Given the evidence suggesting substantial trade and manufacturing at Woodstown, as well as the large number of ship rivets (Russell et al. 2007:15), it seems that a long river frontage may have been a core requirement of the site. The tiny “fort” encountered at The Udal, depending on the circumstances of the Viking settlement in the Western Isles, may never have held a real functional use, especially considering its short life-span, and would certainly be able to accommodate only a small number of people. Could it be here that there was an initial need for small-scale, localized defence, or indeed a display of dominance in the immediate landscape? There is also the possibility that promontory sites were being fortified. At Burton Point, Cheshire, a promontory site is arbitrarily associated with the Viking leader Ingimund (English Heritage 2012b), though such sites are associated with the Viking Age in the Isle of Man (Cubbon 1983, Gelling 1952, Wilson 2008). “Most promontory forts on the island … [are small,] cutting off a headland overlooking the sea by means of a bank and external ditch” (Wilson 2008:18). These sites, however, demonstrate multi-period occupation, and though they certainly seem to have been utilized or reoccupied during the Viking Age, it is unlikely that these had associations with large armed groups such as we would consider an “army”, though it may be that they are related to local elites. Finally, as has been alluded to, it must be noted that fourteen of the sites featured in this study no longer possess extant earthworks that were noted by antiquarian scholars, or the postulated Viking sites have not been associated with any visible archaeological features. With too many variables that could affect the composition of Viking forces, there is probably no diagnostic shape for such a site, which markedly differs from the comparatively easy identification of military fortifications, such as those of the Roman period, for example. What is clear, however, is that there is a significant degree of monumental diversity even in those few sites that have been confidently identified, while other postulated sites take a number of forms. Finally, the extent to how many of these sites (if any) can be considered to really represent Viking fortifications must be considered. Obviously, those revealed through modern archaeological excavation present the most promise, though in the case of The Udal, the relationship between the “fort” and the Scandinavian settlement of the area remains unclear. At Llanbedrgoch, the presence of hacksilver indicates the influence of a silver bullion economy—a strong Scandinavian trait—though it must be considered to what extent the site was used (if at all) by Viking groups; indeed it may have been Viking raiding that led to the eventual abandonment of the site (Redknap 2008:406). The demographics of these sites may also vary; it has been postulated that women found buried in the mass grave at Repton may have been English given that the site was a double house monastery (Biddle and Kjølbye-Biddle 1992, Richards 2003). These monasteries were often established under royal patronage and served the important function of maintaining the cult of the royal family and its members (Foot 2000:336). Isotope analysis on one of the female skulls at Repton, however, suggests that she may have originated from Scandinavia or continental Europe (Budd et al. 2004:138), throwing into question the notion that all 18% of women who made up the mass grave at Repton were English (McLeod 2011:346). Processes such as textile production, which has been revealed at Woodstown, are thought by Russell et al. (2007:37) to be “associated with women rather than men” in the Viking Age and therefore further suggests the presence of women in fortifications, though whether gender roles can realistically be so rigorously applied is open to debate. Some sites, such as the “Aldewerke” in Great Shelford, Cambridgeshire, demonstrate potential, but a lack of datable finds prevents conclusive interpretation at this particular site. This site is taken, however, as being consistent with “the aceramic nature of many Saxon and Danish sites” (Malim 2000:12). Other sites, such as Danbury, Essex, have been so heavily truncated by processes such as market gardening that ever being able to assess their potential may be difficult. Other, unexcavated sites are altogether harder to interpret. The Blunham site possesses some good circumstantial evidence 2013 B. Rafffield 17 (see Edgeworth 2008, Edgeworth and Fradley 2009) but requires excavation of potential archaeological features, such as a curvilinear feature that is possibly a large enclosure or fortification ditch (Fig. 11), if we are to understand it. The northwest corner of the Benfleet fortification may be identified by the grounds of Benfleet church, though the majority of the fortification may likely lie under modern housing and a substantial car park. Other sites have been re-interpreted since being designated as “Viking”; Beeston Berrys, Bedfordshire, for example now being considered to be a deserted medieval village. Other hypothesized “Viking” sites based on the association of a site with stray finds such as Frogmore Lodge, Hertfordshire or Gainsborough, Lincolnshire should certainly be considered at this time for what they likely are—undated earthworks, perhaps related to find boundaries or agricultural activity, or sites spuriously associated with the Scandinavian presence in England through the location of “supporting” artifacts. Others recorded within a general locality or region, such as that recorded in the Stovin Manuscript in the East Riding of Yorkshire, are unlikely ever to be located at all if they even existed. Placing these sites in an international context may assist in establishing possible “norms” for Viking Age fortification sites from which more thorough and constructive arguments for postulated fortification sites can be formed. Being a people with a wide geographical reach, the Vikings can only fully be understood when their actions in Britain are compared with their actions in the wider Viking world. The Study of Fortification Sites in Ir eland and Scandinavia It is necessary to consider Viking Age fortifications that have been subject to investigation outside of Britain. Though they date from various points spanning the entirety of the Viking Age in Britain and existed within individual socio-economic contexts, they still form a body of better understood comparative evidence which can influence how we view the British sites discussed above. As has been mentioned, a number of longphuirt have been located in Ireland (for these and Scandinavian fortification sites discussed, see Fig. 12), of which Dunrally, Athlunkard, and Woodstown consist of D- or B-shaped earthworks, while that at Annagassan is formed by an enclosed peninsula (Linn Duachaill 2012). All are situated by waterways. The mixed nature of the finds at Woodstown Figure 11. The site of the “Castle” at Blunham, Bedfordshire. In the middle distance a curvilinear feature (marked by red arrow), possibly representing a large ditch, can be seen. Photograph by the author . 18 Journal of the North Atlantic No. 20 2013 B. Rafffield 19 suggests combined “military”, “domestic”, and “productive” activities—the latter represented by textile production and silver working—which throws the nature of this site into uncertainty. Evidence suggests that despite Woodstown being enclosed in defences, which may have incorporated palisades (Russell et al. 2007:32), the wide entrance to the site (approximately 7.5 m) would have been an impractically large gap to defend in any fortification. Truncation by 19th-century railway works has obliterated any trace of riverside fortifications, though possible riverside earthworks have been noted at Edgeworth’s site at Blunham on the River Great Ouse (Edgeworth and Fradley 2009). The defensive capacity of Woodstown is also thrown into question due to the relative slightness of the ditches, which measure 1.2 m deep and 3–4 m wide (Russell et al. 2007:32) in comparison to Repton’s 4.2 m deep and 8.5 m wide ditches. Considering those at Dunrally, however, which measure 1.8 m deep and 5.3 m wide (Kelly and Maas 1995:31), the Woodstown ditches may not be significantly smaller in terms of the size of fortifications in Ireland. The activities at the site certainly conform with those postulated to be taking place at Torksey, Lincolnshire, which shows that the juxtaposition between military and productive sites is not limited to Irish sites. The possible double enclosure or B-shape plan of Woodstown may suggest that the site was divided between military and non-military activities, with “domestic” finds being recovered more frequently from the north end of the site, though the partially excavated nature of the site may be responsible for this perceived distribution. With work now being undertaken at Annagassan, County Louth, it may be possible to begin formulating arguments as to what processes were taking place at the Irish sites during the Viking Age. Known fortifications in Scandinavia are surprisingly few in number given the frequent conflict that took place to bring about the formation of the Scandinavian states. The majority of fortifications occur relatively late in terms of the Viking Age in Britain. Alongside the Trelleborg-type fortresses, the settlements of Birka, Sweden and Hedeby, Jutland (now in northern Germany) are notable for the presence of fortifications, the latter being part of an extensive defensive system involving the use of the linear Danevirke and Kovirke earthworks. While a general lack of fortified sites in Scandinavia has been noted by Hedenstierna-Jonson et al. (in press:1), the authors stress that this should not “falsely be interpreted as signs of a less violent time”. The island town of Birka, Sweden was an important center of trade and commerce during the Viking Age. The town is dominated by a hillfort defended by ramparts, which was in use from the mid-8th and mid-9th centuries until the late 10th or early 11th century (Hedenstierna-Jonson 2006:49). This site was therefore used throughout much of the time that the Scandinavians were operating in Britain; thus, it is possible that hillfort sites in Britain could have potential for Viking Age occupation if Scandinavian groups continued to make use of this tradition. It should be noted here, however, that the Birka fortification was specifically a defensive site intrinsically associated with the town itself and demonstrates exceptional use of this fortification type during the Viking Age (Hedenstierna-Jonson 2006:49), with the use of hillfort sites having decreased dramatically in the 6th century. The site therefore differs in nature and function from the fortifications and camps in Britain that this study is concerned with. The Trelleborg-style fortresses surpass any hypothesized Viking fortification sites in Britain both in terms of complexity of construction and uniformity. Comparative examples should again not necessarily be looked for in Britain given the specific socio-political, military, and economic situation within which the Trelleborg fortresses were built during the late 10th century. Indeed, they must also be considered for their function as centers of royal power alongside that of a military establishment. As Dyer (1972) shows, however, linking hypothesized sites with these fortifications does occur. These sites make varying use of stone construction, though defensive banks are consistently constructed of earth and turf (Roesdahl 1986:215) and use an “extraordinary amount of timber … [the rampart] was not only covered by a plank palisade on both sides, but also traversed in its interior, both longwise and clockwise” (Nørlund 1948:273). The fortifications of Fyrkat and Aggersborg, seem strategically placed to monitor transport routeways. Aggersborg, for example, is thought to be situated to monitor the Limfjord, a quick and safe route allowing shipping to traverse the Jutland peninsular and move from the North Sea to the Baltic (Roesdahl 1986:225), respectively. Rosedahl (1986:218) suggests that the striking resemblance of the Trelleborg fortresses to the geometrical fortress at Souburg in the Netherlands may have been inspired by other sites on the continent. Indeed, there is a growing number of possible Viking Age fortifications on the European continent, with those in the Low Countries postulated as refuges constructed against Viking attack, though these date from the last quarter of the 9th century (Van Heeringen 1996:261). Similarly to Britain, the population structure of these fortifications is also Figure 12 (opposite page). Maps displaying Irish and Scandinavian sites discussed in this paper, including additional “Trelleborg”-type fortresses. 20 Journal of the North Atlantic No. 20 thrown open to question by the excavation of women and children in cemeteries at Fyrkat and Trelleborg, though whether they lived within the fortification itself or as part of the wider population outside of the fortresses is unknown. The settlement of Hedeby was enclosed in defenses during the early stages of the 10th century, with the settlement having existed as undefended for some time (Hilberg 2008:106). Substantial evidence for trade and the multicultural population structure seen in reference to the burials there suggests that this site certainly was not primarily of military use, though an attempt was made to bring Hedeby into the wider defensive landscape by the construction of the Kovirke in the 10th century. Evidence for gateways in this structure suggest that the monitoring of traffic into and out of the area was of importance (Dobat 2008:42). It is again necessary to highlight here that comparative equivalents for Hedeby should not be sought in Britain in the terms of this study— Hedeby was a significant settlement and trading center that warranted royal protection and as such directly contrasts with the fortifications that this study aims to address. It is clear, therefore, that the fortifications encountered outside of Britain are of a different character than those that we would expect to find. The Irish longphort sites provide the closest comparison, though whether at this stage they can be truly considered comparable in terms of form and function is subject to opinion. The evidence from Torksey tentatively suggests so, with further investigation both here and at other sites being likely to shed light on this. What can be seen, however, is that once again there exists a considerable amount of monumental diversity within Scandinavia, though regional similarity does exist. As with the Trelleborg fortresses, however, this similarity is due to the royal building programs that initiated their construction and should not be mistakenly taken as a model which we might expect to locate in Britain. Unlike the kind of sites that we might expect in Britain, the Scandinavian fortifications discussed above all involved the investment of significant time, income, and the centralized control of resources—something that is unlikely to have been available to Viking raiding armies while in Britain. Furthermore, these fortifications did not exist in isolation; the example of Hedeby demonstrates that large-scale systems of defense did exist in Viking Age Scandinavia. While it is nonetheless important to consider parallels between the two regions, the differing political, military, and socio-economic contexts attached to the fortifications within them must also be remembered before attempting to pos - tulate associations. At the same time, however, the British sites should not be considered in isolation. Hedenstierna-Jonson et al. (in press) argue that mental boundaries would have been of equal importance to physical ones, though these began to be physically visualized in the landscape through the construction of defensive systems. The importance of topographical boundaries in Scandinavia may be reflected in Viking Age England—the “Treaty of Wedmore”, agreed between Alfred of Wessex and the Viking leader Guthrum in ca. 878 made extensive use of the rivers Thames, Lea, and Great Ouse to divide the south and east of England between Wessex and the Vikings who were to settle in East Anglia. In both the 9th- and 11th- century phases of Viking incursions, the mobile nature of the Viking forces also meant that linear boundaries were largely rendered obsolete. The topography of Denmark when compared to the rest of Scandinavia may explain the greater number of fortifications and linear earthworks there. The Danish landscape itself was perhaps less of a defense, as large, sparsely populated areas of land that exist in other parts of Scandinavia effectively allowed threats to be absorbed by the landscape and countered before reaching important population centers. The substantial fortifications at Birka were most probably needed due to the relatively easy access by which the settlement could be approached by water and its enviable placement at a junction of trade routes. It doses seem likely, however, that Birka was a the center of an extensive defense-in-depth system that existed in the Lake Mälar region (Hedenstierna-Jonson et al., in press). The presence of possible interior mounds at postulated British and Irish sites could represent the location of central structures of importance. Hedenstierna-Jonson et al. (in press:11–12) state that in Viking Age Scandinavia it was not was the castle or fort, but the hall that held great importance. These replaced the hillforts of the Migration Period, and not until “sometime well into the Middle Ages [were] the great halls … replaced by citadels and castles combining the representative and symbolic functions of the halls, and hilltop sites before them, with more advanced military capacities.” The incorporation of a hall into any site, however, would certainly suggest at least a semi-permanence on the part of the garrison and the fortification itself. In light of the evidence from Scandinavia, we are therefore presented with a series of paradoxes or juxtapositions in terms of Viking fortifications in Britain. While there has evidently been a concern with identifying fortifications in Britain in the past based on Scandinavian sites, these interpretations cannot always be justified. The mobile nature of Viking forces mean that on a day-to-day basis camps would probably have been undefended, with only the semi-permanent overwintering camps being enclosed in defensive earthworks, though it 2013 B. Rafffield 21 is possible that even these were only foci of the Viking forces settled in the wider landscape. If and when fortifications were constructed, we cannot be sure as to what form they took—while shipborne forces may well have utilized D-shaped enclosures to protect their vessels, this is not a purely Viking trait. The mixed and fluctuating numbers in Viking “armies” and the fact that individual leaders would have their own methods of waging war means that we cannot necessarily expect any uniformity in the form, size, or even presence of fortifications. Evidence from Britain, Ireland, and the continent suggests that the Vikings often utilized natural defenses such as waterways, rock outcrops, or islands, the former recorded as being used in 870 at Reading when the “great army” established a camp thought to be between the rivers Thames and Kennet (Youngs et al. 1987), while hillforts are popular sites of refuge and defense in Scandinavia. The familiarity with this type of defensive structure could suggest that extant Iron-Age hillforts, or other prehistoric, Roman, or Anglo-Saxon enclosures or fortifications in Britain were re-occupied by Viking armies both for overwintering (for example York in 866–867, Nottingham in 867–868, and Exeter in 876 [Swanton 2000:68,74]) as well as in the longer term (see the recorded occupations of Colchester prior to 917 [Swanton 2000:102] and Lincoln and Leicester among others until various points in the 10th century). The use of islands, such as those at Mersey and Northey Islands, Essex is attested in the contemporary sources. These locations could provide a fixed geographical limit within which to undertake the study of Viking activities if occupation sites can be located. Conclusions and Suggestions for Future Investigations This study has demonstrated there is, in some cases, a significant antiquarian influence on our interpretations of Viking Age monuments in parts of Britain. The inevitable result of this influence is that these monuments remain very much understudied and enigmatic. Though inroads are now being made into the study of overwintering camps, our knowledge must be dramatically expanded if we are to understand the fortifications of the Viking armies that were operating in Britain, which could also significantly impact our knowledge of how they were (or were not) assimilated into Anglo-Saxon society following the establishment of the “Danelaw”. Examples from Europe and Scandinavia show that potential Viking fortified sites can vary greatly in their location, form, and size, and also in the activities which were taking place within them, something which is also supported by the British sites. While D-shaped fortifications on riverbanks may well be representative of shipborne Viking armies, it is unlikely that sites in a dry-land context would conform to a rigid layout in terms of their defensive plan. With the use of natural defenses minimizing the need for earthworks, the archaeological signature of such sites may not be substantial, and camps of a more transitory nature may not have utilized earthworks at all, instead relying on sentries to watch for an impending attack. As such, the archaeological signature of these camps may be limited to an artifact scatter and evidence of activities such as metal working or commerce, as seems thus far to be the case at Torksey. The contrast between the Irish longphort at Woodstown and the Viking camp at Repton shows that while these sites may be similar in form, the size of defenses may vary depending on the imminence of a military threat or the wish to pursue goals such as the symbolic military dominance of an important site of regional power, though practical issues such as the depth of the water table may also have an impact. Even where defenses have been constructed, these may have been obliterated aboveground over the last millennium, and for this reason, some sites will only be identified through intrusive archaeological investigation. That these sites were primarily defensive in their purpose does not mean that significant insights into daily Viking “domestic” life would not be encountered. With overwintering sites being utilized perhaps for a number of months, it is likely that evidence of activities wholly unconcerned with warfare would be observed. In addition to this, it should not be assumed that Viking armies did not attract the same camp followers that are associated with armies throughout known history, with Repton producing likely evidence of Scandinavian or European females accompanying the “great army” (McLeod 2011). Understanding how these groups interacted in the social and domestic spheres while acting as participants within a general context of conflict could provide the basis for the knowledge of a previously unknown facet of Viking Age life. We should expect some sites to exhibit signs of destruction, as has been noted at the fortification at Camp de Péran, Brittany, where a circular fortification seems to have been either occupied or attacked by Vikings, with the collapsed rampart sealing a coin minted at York in 902–925 (Price 1991:4). Fortifications that were subject to siege such as Benfleet or Tempsford would likely demonstrate similar activity. It is likely, as excavations at Repton, Woodstown, Fyrkat, and Trelleborg show, that burials will be present at fortification sites, though whether these truly represent the populations of these sites is open to debate. While at present no diagnostic indicators for a Viking origin to sites may confidently be relied 22 Journal of the North Atlantic No. 20 upon, it is only through the re-evaluation and detailed study of these sites that we can begin to understand them. Indeed, the Vikings have proved to be “as elusive to us today as they were to their contemporaries” (Clarke 1999:36). Finds from the few excavated sites that have received archaeological attention testify to hitherto unconsidered processes taking place within fortifications which shows that these monuments have the potential to transform our views of the Viking Age. The evidence for nonmilitary activities such as manufacturing, trade, and exchange within a silver bullion economy ironically provides a Scandinavian archaeological signature from which it may be possible to identify fortification and campaign base sites used by Viking armies or warbands. Archaeologists should endeavor to study potential Viking sites within their landscape contexts, as local means of communication such as rivers and roads as well as the presence of local and regional power centers may have been instrumental in the placement of fortifications. By being able to place the Vikings in space as well as time, there is the potential to better understand the military, religious, and socio-political impacts of the Viking Age. It is clear that we must move towards an in-depth and individual approach to these sites and Viking Age warfare in general if we are to better understand this archaeologically important period. Establishing exactly which direction this research should take, however, is difficult. The merits of a deductive approach were considered at the beginning of this paper and determined unsuitable, as this would not take into account local, regional, or national cultural variations and influences. It is for this reason that resources such as the National Mapping Program were not used—there is little use in picking out potential “Viking” sites when we are not sure what exactly to look for. At present, the best approach may be to prioritize and address those sites previously identified or supported through modern scholarship, such as the “Aldewerke” in Cambridgeshire, Edgeworth’s site at Blunham, and the site currently under investigation at Torksey. Through comparison with sites that have previously been excavated (such as Willington Docks and Repton), it might be possible to begin to form ideas of what we might and might not expect to find, allowing afterwards for a deductive approach to fortification sites. If research at Torksey shows that fortifications were not present there, the entire notion of such sites could be thrown into question—is it possible that a substantial army like that camping at Torksey would not even require fortifications to defend itself, especially when it is recorded in contemporary sources as having made peace with the Mercians (Swanton 2000:72)? Could Repton with its substantially defended compound actually be relatively anomalous, its presence being explained through the occupation and domination of a Mercian royal institution by the Viking “great army” thereby legitimizing their claim to Mercia and forcing the exile of King Burhred to the continent? It must also be considered that, while the fortification may be an inner enclosure which existed as part of a much larger enclosure (similar to “citadels” noted at sites such as Dunrally, Ireland), it may also be possible that no fortification was conceived to house the entire “great army” at all, with various groups dispersing throughout the immediate area; the Annals of St. Bertin recorded such a practice on the continent (Lund 1985:112). It may be that Repton was constructed for unique political and military means and should not be taken as an example to look for elsewhere. The use of metal detectors has been proved at Torksey, where hundreds of objects have been recovered, though the preponderance of nonferrous objects clearly shows a need for more intensive survey to locate ferrous material. If we can truly expect trade and commerce to be taking place at such sites, then organized metal-detecting surveys could perhaps be instrumental in locating them. For other fortifications such as Tempsford, however, the fact that this was a “campaign fortification” constructed for the armies of East Anglia and Huntingdon to strike at Bedford may suggest that these processes may not be so evident. The site at Blunham itself would substantially benefit from excavation of features at the site to assist in proving or disproving the postulated association with the 917 fortification. It is fitting to end this paper with a statement from Goddard (1904:308), who, writing at the turn of the 20th century, states that, “the spade is … the agent most in request to let in fresh light on the subject.” Little has changed since; modern, intrusive archaeological investigation is the primary means through which we will continue to illuminate the nature and function of these sites and better understand Viking fortification sites. Only through data gained from excavations will it be possible to construct a model that can be applied to other suspected sites. Once this model is established, it may become truly possible to begin to understand the sites constructed and occupied by Viking armies and therefore better understand their important place in British history. Acknowledgments Thanks go to Neil Price, John Carman, Amanda Wynne, and Laura Whitehouse for reading and commenting on drafts of this paper, with thanks also going to Charlotte Hedenstierna-Jonson, Lena Holmquist, and Michael Olausson for allowing me to utilize their paper on Viking fortifications in Scandinavia. Thanks also to John Carman and Chris Callow at the University of Birmingham for supervising the research from which this work derives. 2013 B. Rafffield 23 Literature Cited Abels, R. 1997. English logistics and military administration 871–1066: The impact of the Viking wars. Pp. 257–265, In A. Nørgård Jørgensen and B. Clausen (Eds.). Military Aspects of Scandinavian Society in a European Perspective, AD 1–1300, (Publications from the National Museum, Studies in Archaeology and History, vol. 2). National Museum, Copenhagen, Denmark. 265 pp. Abels, R. 2003. Alfred the Great, the Micel Hoeðen Here and the Viking Threat. Pp.265–279, In T. Reuter (Ed.). Alfred the Great: Papers from the Eleventh Centenary Conferences. Ashgate, Aldershot, UK. 416 pp. Allcroft, A.H. 1908. Earthwork of England. MacMillan and Co., London, UK. 711 pp. Allen, T. 1834. The History of the County of Lincoln. John Saunders, Lincoln, UK. 360 pp. Armitage, E.S. 1900. Anglo-Saxon Burhs and Early Norman Castles. Proceedings of the Society of Antiquaries of Scotland 34:260–288. Bassett, S. 2007. Divide and rule? The military infrastructure of eighth- and ninth-century Mercia. Early Medieval Europe 15(1):53–85. Beresford, G. 1987. Goltho: The development of an early medieval manor c. 850–1150. Historic Buildings and Monuments Commission for England, London, UK. 218 pp. Biddle, M., and B. Kjølbye-Biddle. 1992. Repton and the Vikings. Antiquity 66:36–51 Biddle, M., and B. Kjølbye-Biddle. 2001. Repton and the “Great Heathen Army”, 873–874. Pp. 45–96, In J. Grahm- Campbell, R. Hall, J. Jesch, and D.N. Parsons (Edss). Vikings and the Danelaw: Select Papers from the Proceedings of the Thirteenth Viking Congress. Oxbow, Oxford, UK. 368 pp. Blackburn, M. 2002. Finds from the Anglo-Scandinavian site of Torksey, Lincolnshire. Pp. 89–101, In B. Paszkiewicz (Ed.). Moneta Mediævalis: Studia numizmatyczne I historyczne ofiarowane Profesorowi Stanislawowi Suchodolskiemu w 65. rocznicę urodzin, Wydawn DiG, Warsaw, Poland. 571 pp. Blackburn, M. 2011.Viking Coinage in the British Isles. Spink, London, UK. 416 pp. Boyd Dawkins, W. 1873. On some human bones found at Buttington, Montgomeryshire. The Montgomeryshire Collections 6:141–145. Brooks, N. 1964. Unidentified forts of the Burghal Hidage. Medieval Archaeology 8:74–90. Brooks, N. 1979. England in the ninth century: The crucible of defeat. Transactions of the Royal Historical Society 1979:1–20. Bryant, G. 1994. The Early History of Barton Upon Humber. Barton WEA, Barton, UK. 168 pp. Budd, P., A. Millard, C. Chenery, S. Lucy, and C. Roberts. 2004. Investigating population movement by stable isotope analysis: A report from Britain. Antiquity 78:127–141. Cade, J. 1785. Conjectures concerning come undescribed Roman Roads, and other Antiquities in the County of Durham. In a Letter to the Rev. Dr. Kaye (noe Dean of Lincoln) from John Cade, Esq. of Durham. Archaeologia 7:74–81. Chalkley Gould, I. 1903. Ancient Earthworks. Pp. 275–313, In H.A. Doubleday and W. Page (Eds.). The Victoria History of the Counties of England: A History of Essex, Vol. 1. Archibald Constable, London, UK. 598 pp. Chalkley Gould, I. 1908. Ancient earthworks. Pp. 389– 445, In W. Page (Ed.). The Victoria History of the County of Kent, Vol. I. Archibald Constable, London, UK. 622 pp. Chippindale, C. 1983. Stonehenge Complete. Thames and Hudson, London, UK. 296 pp. Chippindale, C. 1994. Stonehenge Complete (Rev. Ed.). Thames and Hudson, London, UK. 296 pp. Christiansen, E 2002. The Norsemen in the Viking Age. Blackwell, Oxford, UK. 392 pp. Clark, G.T. 1884. Medieval Military Architecture in England Vol. II. Wyman and Sons, London, UK. 548 pp. Clarke, H.B. 1999. The Vikings. Pp. 36–58, In M. Keen (Ed.). Medieval Warfare. Oxford University Press, Oxford, UK. 340 pp. Cohen, S. 1965. Viking Fortresses of the Trelleborg Type. Rosenkilde and Bagger, Copenhagen, Denmark. 103 pp. Crawford, I., and R. Switsur. 1977. Sandscaping and C14: The Udal, N. Uist. Antiquity 51:124–36. Cubbon, M. 1983. The archaeology of the Vikings in the Isle of Man. Pp. 13–26, In C. Fell P. Foote, J. Graham- Campbell, R. Page, and D.M. Wilson. (Eds.). The Viking Age in the Isle of Man. Select papers from the Ninth Viking Congress, Isle of Man, 4–14 July 1981. Viking Society for Northern Research, London, UK. 187 pp. Davison, B.K. 1972. The Burghal Hidage Fort of Eorpeburnan: A suggested identification. Medieval Archaeology 16:123–7. Dobat, A.S. 2008. Danevirke revisited: An investigation into military sociopolitical organisation in South Scandinavia (c. AD 700 to 1100). Medieval Archaeology 52:27–67. Downham, C. 2008. Vikings in England. Pp. 341–349, In S. Brink and N. Price (Eds.). The Viking World. Routledge, London, UK and New York, NY, USA. 717 pp. Dyer, J. 1962. Danish Earthworks. Bedfordshire Magazine 8:347. Dyer, J. 1972. The earthworks of the Danelaw frontier. Pp. 222–236, In P.J. Fowler (Ed.). Archaeology and the Landscape: Essays for L.V. Grinsell. John Baker, London, UK. 263 pp. Edgeworth, M. 2006. Notes on a previously unknown “castle” near Tempsford: A historic battlefield. Unpublished. [Available from the author at University of Leicester, UK]. Edgeworth, M. 2008. The Tempsford Project: An interim report. South Midlands Archaeology 38:8–16. Edgeworth, M., and M. Fradley. 2009. Investigations in Tempsford 2009. Unpublished. [Available from the first author at University of Leicester, UK]. Edmondson, G., and A. Mudd. 2004. Medieval occupation at “Danish Camp”, Willington. Bedfordshire Archaeology 25:208–221. English Heritage. 2007. Canute’s camp. Available online at http://www.pastscape.org.uk/hob.aspx?hob_ id=419103. Accessed 23August 2011. 24 Journal of the North Atlantic No. 20 English Heritage. 2011. List entry summary. Available online at http://www.heritagegateway.org.uk/Gateway/ Results_Single.aspx?uid=1017206andresourceID=5. Accessed 14 September 2011. English Heritage. 2012a. Manwar Ings. Available online at http://www.pastscape.org.uk/hob.aspx?hob_id=352 577&sort=4&search=all&criteria=manwar%20ings& rational=q&recordsperpage=10. Accessed 24 August 2012. English Heritage. 2012b. Promontory fort on Burton Point 55 m SW of Burton Point Farm. Available online at http://www.heritagegateway.org.uk/Gateway/Results_ Single.aspx?uid=MCH6214&resourceID=1004. Accessed 16 November 2012. Foot, S. 2000. Nunneries. Pp. 336, In M. Lapidge, J. Blair, S. Keynes, and D. Scragg (Eds.). The Blackwell Encyclopaedia of Anglo-Saxon England. Blackwell, Oxford, UK. 560 pp. Forte, A., R. Oram, and F. Pedersen. 2005. Viking Empires. Cambridge University Press, Cambridge, UK. 447 pp. Gaffney, V., and J. Gater. 2003. Athelney (II) Somerset. 2002/59. Geophysical Surveys of Bradford (GSB Prospection), Bradford, UK. Gazin-Schwartz, A., and C. Holtorf (Eds.). 1997. The Archaeology of Folklore. Routledge, London, UK. 304 pp. Gelling, P.S. 1952. Excavations of a promontory fort at Port Grenaugh, Isle of Man. Proceedings of the Isle of Man Natural History and Antiquarian Society 5(3):307–15. Gibbons, M. 2004. The longphort phenomenon in early Christian and Viking Ireland. History Ireland 12(3):19–23. Goddard, A.R. 1902. The Danish camp on the Ouse near Bedford. Saga Book of the Viking Club 3:331–337 Goddard, A.R. 1903. The Danish attack on Bedford in 921. The Bedfordshire Times 27 March 1903. Goddard, A.R. 1904. Ancient earthworks. Pp. 267–308, In H.A. Doubleday (Ed.). The Victoria History of the County of Bedfordshire I. Archibald Constable, London, UK. 598 pp. Graham-Campbell, J., and C. Batey. 1998. Vikings in Scotland: An Archaeological Survey. Edinburgh University Press, Edinburgh, UK. 288 pp. Gray, H. 1933. Trial-excavations in the so-called “Danish Camp” at Warham, near Wells, Norfolk. The Antiquaries Journal 13:399–413. Griffith, P. 1995. The Viking Art of War. Casemate, London, UK. 244 pp. Griffiths, B. 1991. The Battle of Maldon: Text and Translation. Anglo-Saxon Books, Pinner, UK. 89 pp Hall, R. 2007. Exploring the World of the Vikings. Thames and Hudson, London, UK. 240 pp. Harris, J. 1719. History of Kent. Publisher unknown. Hart, C. 1995. The Aldewerke and Minster at Shelford, Cambridgeshire. Pp. 42–68, In D. Griffiths (Ed.). Anglo-Saxon Studies in Archaeology and History 8. Oxford University Press, Oxford, UK.143 pp. Haslam, J. 2005. King Alfred and the Vikings: Strategies and tactics 876–886 AD. Pp. 121–153, In S. Semple (Ed.). Anglo-Saxon Studies in History and Archaeology 13. Oxbow, Oxford, UK. 368 pp. Haslam, J. In press (a). Planning in late Saxon Worcester. Draft available online at http://jeremyhaslam.wordpress. com/. Accessed 20 July 2012. Haslam, J. In press (b). A probable late Saxon burh at Ilchester. Draft available online at http://jeremyhaslam. wordpress.com/. Accessed 20 July 2012. Hassall, T. 1975. Excavations at Willington 1973. Bedfordshire Archaeological Journal 10:25–40. Hasted, E. 1782. The History and Topographical Survey of the County of Kent: Volume 6, W. Bristow, Canterbury, UK. 498 pp. Available online at http://www. britishhistory.ac.uk/report.aspx?compid=62959. Accessed 8 June 2009. Hedenstierna-Jonson, C. 2006. The Birka warrior: The material culture of a martial society. Ph.D. Dissertation. Stockholm University, Stolkhom, Sweden. 118 pp. Hedenstierna-Jonson, C., L. Holmquist, and M. Olausson. In press. The Viking Age Paradox: Continuity and Discontinuity of Fortifications and defense Works in Eastern Scandinavia. Proceedings from Landscapes of defense in the Viking Age. Brepols, Turnhout, Belgium. Heritage Gateway. 2006. Sub-circular enclosure (possible Danish burh), East Acridge. Available online at http:// www.heritagegateway.org.uk/Gateway/Results_Single. aspx?uid=MLS17906andresourceID=1034. Accessed 6 March 2012. Hill, D., and A. Rumble (Eds.). 1996. The defense of Wessex: The Burghal Hidage and Anglo-Saxon Fortifications. Manchester University Press, Manchester, UK. 256 pp. Hilberg, V. 2008. Hedeby: An outline of its research history. Pp. 101–111, In S. Brink and N. Price (Eds.). The Viking World. Routledge, London, UK and New York, NY, USA. 717 pp. Jackson, C. 1882. The Stovin manuscript. Yorkshire Archaeological and Topographical Journal 7:201–202. Jankuhn, H. 1933. Die Ausgrabungen in Haithabu 1930– 1933. Nordelbingen 9:341–369. Kelly, E.P., and J. Maas. 1995. Vikings on the barrow: Dunrally Fort, a possible Viking longphort in County Laois. Archaeology Ireland 9(3):30–32. Kelly, E.P., and J. Maas. 1999. The Vikings and the kingdom of Laois. Pp. 123–59, In P.G. Lane and W. Nolan (Eds.). Laois. History and society. Geography Publications, Dublin, Ireland. 748 pp. Kelly, E.P., and E. O’Donovan. 1998. A Viking longphort near Athlunkard, Co. Clare. Archaeology Ireland 12(4):13–16. Kilburn, R. 1659. Topographie or Survey of the County of Kent. H. Atkinson, London, UK. 422 pp. Kitchen, F. 1984. The Burghal Hidage: Towards the identification of Eorpeburnan. Medieval Archaeology 28:175–178. Knorr, F. 1924. Schleswid un Haithabu. Shleswig-Holsteinisches Jahrbuch für 1924 (Schleswig Heimatbuch I):24–31. Kramer, W. 1999. Neue untersuchungen im Hafen von Haithab. Archäologische Nachrichten au Schleswig- Holstein 9/10(1998/9):90–118. Laing, L., and J. Laing. 1985. The Dark Ages of West Cheshire. Cheshire County Council, Chester, UK. 67 pp. 2013 B. Rafffield 25 Olsen, O. 1977. Fyrkat: En jysk vikingeborg, København, Det kgl. nordiske Oldskriftselskab: i kommission hos H.H.J. Lynge and Søn, Copenhagen, Denmark. 282 pp. Piggot, S. 1989. Ancient Britons and the Antiquarian Imagination. Thames and Hudson, London, UK. 175 pp. Pitts, L.F., and J.K.S. St. Joseph. 1985. Imchtuthil: The Roman Legionary Fortress. Excavation: 1962–65. The Society for the Promotion of Roman Studies, London, UK. 344 pp. Pollard, H.P. 1906. Danesfield, Watton. Transactions of the East Hertfordshire Archaeological Society 3:176. Price, N. 1991. Viking armies and fleets in Brittany: A case study for some general problems. Pp. 7–24, In H. Bekker-Nielsen and H. Frede Nielsen (Eds.). Beretning fra tiende tværfaglige Vikingesymposium. Odense University Press, Odense, Denmark. 56 pp. Prior, C.E. 1886. Unknown document referenced in Bedfordshire HER documents. Raffield, B. 2009. “Plundering the territories in the manner of the heathens”: Identifying Viking Age battlefields in Britain. Rosetta 7:22–43. Available online at http:// rosetta.bham.ac.uk/issue7/plundering-territories/. Redknap, M. 2000. The Vikings in Wales: An Archaeological Quest. National Museums and Galleries Wales, Cardiff, UK. 116 pp. Redknap, M. 2007. Llanbedrgoch: An early medieval settlement and its significance. Transactions of the Anglesey Antiquarian Society 48:53–72. Redknap, M. 2008. The Vikings in Wales. Pp. 401–410, In S. Brink and N. Price (Eds.). The Viking World. Routledge, London, UK and New York, NY, USA. 717 pp. Redknap, M. 2009. Silver and commerce in Viking-Age North Wales. Pp. 29–44, In J. Graham-Campbell and R. Philpott (Eds.). The Huxley Viking Hoard: Scandinavian Settlement in the North West. National Museums Liverpool, Liverpool, UK. 94 pp. Reynolds, A. 1999. Later Anglo-Saxon England: Life and Landscape. Tempus, Stroud, UK. 208 pp. Reynolds, A. 2003. Boundaries and settlements in later 6th to 11th century England. Pp. 97–139, In D. Griffiths, A. Reynolds, and S. Semple (Eds.). Boundaries in Early Medieval England, Anglo-Saxon Studies in Archaeology and History 12, Oxford University Press, Oxford, UK. 217 pp. Richards, J. 2003. Pagans and Christians at the frontier: Viking burial in the Danelaw. Author manuscript. Available online at http://eprints.whiterose.ac.uk/archive/ 00000755/. Accessed 14 June 2011. Richards, J. 2004. Viking Age England. Tempus, Stroud, UK. 254 pp. Rodwell, W., and C. Atkins. 2011. St Peter’s, Barton Upon Humber, Volume 1: History, Archaeology and Architecture. Oxbow Books, Oxford, UK. 944 pp. Roesdahl, E. 1977. Aggersborg. The Viking settlement and fortress. Chateau Gaillard VIII:269–278. Roesdahl, E. 1986. The Danish geometrical Viking fortresses and their context. Anglo-Norman Studies IX: Proceedings of the Battle Conference, 1986:208–226. Roesdahl, E. 1998. The Vikings, 2nd Edition. Penguin, London, UK. 268 pp. Rushworth, A. 2009a. Housesteads Roman Fort – The grandest station. Volume 1: Structural Report and Discussion. English Heritage, Swindon, UK. 338 pp. Lane, A. 2007. Ceramic and cultural change in the Hebrides AD 500–1300. Cardiff Studies in Archaeology Specialist Report Number 29, Cardiff School of History and Archaeology, Cardiff, UK. Lavelle, R. 2010. Alfred’s Wars: Sources and Interpretations of Anglo-Saxon Warfare in the Viking Age. Boydell Press, Woodbridge, UK. 378 pp. Laver, H. 1903. Benfleet. Transactions of the Essex Archaeological and Historical Society 8:233–236. Laver, P.G. 1930. Pandal Wood Camp. Transactions of the Essex Archaeological and Historical Society 19:255–259. Linn Duachaill. 2012. Home page. Available Online at http://www.linnduachaill.ie/. Accessed 20 July 2012. Livingston, M. (Ed.). 2011. The Battle of Brunanburh. A Casebook. University of Exeter Press, Exeter, UK. 441 pp. Logan, D. 2005. The Vikings in History 3rd Edition..Routledge, London, UK and New York, NY, USA. 205 pp. Lund, N. 1985. The armies of Swein Forkbeard and Cnut: Leding or Lið? Anglo-Saxon England 15:105–126. Lysons, Rev. D. 1796. The Environs of London. T. Cadell, London, UK. 609 pp. Maas, J. 2008. Longphort, Dún, and Dúnad in the Irish annals of the Viking period. Peritia 20:257–275. Malim, T. 2000. (Ed.). An Earthwork Enclosure at Grantham’s Farm, Great Shelford, Cambridge (TL 465 532). An Archaeological Evaluation. Cambridge County Council Archaeological Field Unit, Cambridge, UK. Unpublished. Marrat, W. 1816. The History of Lincolnshire, Boston. W. Marratt, Boston, UK. 153 pp. McLeod, S. 2006. Feeding the mice here in England, c. 865–878. Journal of the Australian Early Medieval Association 2:141–156. McLeod, S. 2011. Warriors and women: The sex ratio of Norse migrants to eastern England up to 900 AD. Early Medieval Europe 19(3):332–353. Medlycott, M. 1993. Danbury camp. Essex Archaeology and History 24:205. Medlycott, M. 2003. Canewdon historic settlement assessment. Essex County Council, Chelmsford, UK. Morley, C. 1924. The sea port of Frostenden. Proceedings of the Suffolk Institute of Archaeology and Natural History 18(3):167–179. Morris, S., and D.G. Buckley. 1978. Excavations at Danbury camp, Essex, 1974 and 1977. Essex Archaeology and History 10:1–28. Musson, C.R., and C.J. Spurgeon. 1988. Cwrt Llechrhyd Llanelwedd: an unual moated site in Central Powys. Medieval Archaeology 32:97–109. Nicolardot, J-P. 1984. Camp de Péran, Plédran (Côtes du Nord). Rapport scientifique des fouilles executes en 1983, Plédran (Côtes-du-Nord). Association des Amis du Camp de Péran 12. Nicolardot, J-P., A. Nissen-Jaubert, and W.H. Wimmers. 1987. Plédran (Côtes du Nord), Camp de Péran. Pp. 229–231, In X. Barral I Altet (Ed.). Le Paysage Monumental de la France autour de l’an mil. Picard, Paris, France. 797 pp. Nørlund, P. 1948. Trelleborg. Bekostet af Carlsbergfondet. København: I Kommission hos Gyldendalske Boghandel, Nordisk Forlag, Copenhagen, Denmark. 296 pp. 26 Journal of the North Atlantic No. 20 Williams, G. 2008. Raiding and Warfare. Pp. 193–203, In S. Brink and N. Price (Eds.). The Viking World. Routledge, London, UK and New York, NY, USA. 717 pp. Williams, G. In press. Viking Warfare and Military Organisation. London, UK. Williams, S. 1912. Renhold. Pp. 214–218, In W. Page (Ed.). The Victoria History of the Counties of England: A History of Bedfordshire Vol. III. Archibald Constable, London, UK. 462 pp. Wilson, D.M. 2008. Vikings in the isle of Man. Aarhus University Press, Aarhus, Denmark. 156 pp. Woodard, P. 1986. Letter to Bruce-Mitford, R. regarding excavations at South Cove. Unpublished.Excavation archive presented to Southwold Museum, Suffolk, UK. Youngs, S.M. 1982. Medieval Britain and Ireland in 1981. Medieval Archaeology 26:164–227. Youngs, S.M., J. Clark, and T. Barry. 1987. Medieval Britain and Ireland in 1986. Medieval Archaeology 31:110–191. Rushworth, A. 2009b. Housesteads Roman Fort – The Grandest Station. Volume 2: The Material Assemblages. English Heritage, Swindon, UK. 614 pp. Russell, I.R., S.H. Harrison, J. Nicholls, J. Kinsella, S. McNamara, and M. O’Hare. 2007. Woodstown 6 supplementary research report. Unpublished Report, Archaeological Consultancy Services Limited, Drogheda, Ireland. Sawyer, P.H. 1971. The Age of the Vikings (2nd Edition). Edward Arnold, London, UK. 275 pp. Schietzel, K. 1981a. Stand her siedlungsarchäologischen Forschung in Haithabu – Ergebnisse und Probleme. In K. Schietzel (Ed.). Berichte über die Ausgrabungen in Haithabu, Vol. 1. Wachholtz, Neumünster, Germany. 98 pp. Schietzel, K. (Ed.). 1981b. Berichte über die Ausgrabungen in Haithabu, Vol. 1. Wachholtz, Neumünster, Germany. 98 pp. Schnapp, A. 1996. The Discovery of the Past. The British Museum, London, UK. 387 pp. Semple, S. (Ed.). 2005. Anglo-Saxon Studies in History and Archaeology 13. Oxbow, Oxford, UK. 368 pp. Sheehan, J. 2008. The longphort in Viking Age Ireland. Acta Archaeologica 79:282–295 Siddorn, K. 2005. Viking Weapons and Warfare. Tempus, Stroud, UK. 192 pp. Smith, C.E. 2008. Buttington, Welshpool. Archaeology Wales 48:110–11. Spurrell, F.C.J. 1885. Early sites and embankments on the margins of the Thames Estuary. The Archaeological Journal 43:269–30.3 Spurrell, F.C.J. 1890. Shoebury Camp, Essex. The Archaeological Journal 47:78–81. Stark, A. 1817. History and Antiquities of Gainsborough. Longman, Orne and Company, Gainsborough, UK. 630 pp. Sutton, D. 2004. Essex. Translation of Camden, W. 1607. Britannia. Available online at http://www.philological. bham.ac.uk/cambrit/essexeng.html#essex1. Accessed 12 June 2009. Swanton, M. 2000. The Anglo-Saxon Chronicles. Phoenix Press, London, UK. 363 pp. Thompson, P. 1856. The History and Antiquities of Boston. John Noble Jun, Boston, UK. 823 pp. Trigger, B.G. 2006. A History of Archaeological Thought, 2nd Edition. Cambridge University Press, Cambridge, UK. 710 pp. University College London (UCL). 2005. Beyond the Burghal Hidage: Anglo-Saxon Civil defense in the Viking Age. Available online at http://www.ucl.ac.uk/ archaeology/research/projects/burghalhidage. Accessed 6 March 2012. University of York. 2012. Viking Torksey. Available online at http://www.york.ac.uk/archaeology/research/ current-projects/torksey/. Accessed 7 March 2012. Van Heeringen, R. 1996. Zeeland Between Nehalennia and the Goth. Pp. 255–262, In M. Lodewijckx (Ed.). Archaeological and Historical Aspects of West- European Societies. Leuven University Press, Leuven, Belgium. 516 pp. Wheeler, W.H. 1896. A History of the Fens of South Lincolnshire. J.M. Newcomb, Boston, UK. 489 pp. 2013 B. Rafffield 27 Appendix 1. Possible Viking fortification sites in the UK located as part of the data search. Site noted in contemporary County/site name Modern source(s) Original source(s) annal/chronicle? Other evidence England Bedfordshire Willington Docks Dyer (1972), Goddard (1902), No Goddard notes “nausts” and a “harbor” Hassall (1975), Goddard (1904) —features suggesting a Danish Edmondson and “waterburg”. Excavation shows this Mudd (2004) to be a later medieval site. Clapham Dyer (1972) N/A No None - site is largely destroyed by gravel digging. Church Panel Dyer (1962) Goddard (1904) No None Dyer (1972) Manor Farm Dyer (1962) Goddard (1904) No None Dyer (1972) Etonbury Dyer (1972) Goddard (1904) No None Renhold Dyer (1972) Prior (1886) No None Goddard (1904) Williams (1912) Seymour’s Mount Dyer (1972) Goddard (1904) No None Risinghoe N/A Goddard (1904) No None Gannocks Castle Dyer (1972) Goddard (1903) Yes (thought to represent Goddard states that a mound covering the 917 Tempsford the entrance in the south east of the fortification) - ASC enclosure is a typically Danish feature. annal A Now thought to be a later medieval site. Beeston Berrys Dyer (1972) N/A Yes (thought to represent None. Has since been reclassified as a the 917 Tempsford deserted medieval village. fortification) - ASC annal A Tempsford/Blunham Edgeworth (2006, Camden’s reference Yes (thought to represent The site is situated at a confluence of 2008), to “Temesford” the 917 Tempsford rivers enclosing a length of Edgeworth and (1607) fortification) - ASC annal A shoreline. Cropmarks show a Fradley 2009 (in Sutton 2004) D-shaped ditch enclosing the site, which may be situated within a much larger D-shaped enclosure. Camden refers to a fort at “Temesford" in Britannia, although this most probably simply demonstrates a knowledge of the Anglo-Saxon Chronicle Berkshire Danish “camp” Youngs et al. N/A Yes – Asser’s Life of King None - suggested that this is between Rivers (1987) Alfred (35) nearer the confluence of the Thames and Kennet Kennet and Thames. Cambridgeshire The “Aldewerke”, Malim (2000), N/A No Possible aceramic period between Great Shelford Hart (1995) Roman and Norman period—Viking occupation? Cheshire Burton Point Laing and Laing N/A No Similar sites exist on the Isle Promontory Fort (1985) of Man, some of which appear to have been constructed or occupied by Vi kings. Derbyshire “Viking Camp” at Biddle and N/A Yes (though a fortification Burials dating to the 10th and 11th Repton Kjølbye-Biddle is not specifically centuries were cut into the ditch (1992,2001) mentioned) – ASC A, E, fill. Also, the discovery of a large Asser’s Life of King Alfred number of skeletons with Viking (46) artifacts (e.g., Thor’s Hammers). Some of the skeletons also displayed possible signs of battle trauma. 28 Journal of the North Atlantic No. 20 Site noted in contemporary County/site name Modern source(s) Original source(s) annal/chronicle? Other evidence County Durham “Danish Camp” near. N/A Cade (1785) No None Sunderland Bridge. Mainsforth, Nab Hill East Riding of Yorkshire Twin Rivers N/A Stovin (1752) in No Camp supposedly located somewhere Jackson (1882) in the parish of Twin Rivers. Essex Butts Hill, Canewdon, Medlycott (2003) Chalkley Gould No Old map evidence shows an oblong Rochford (1903) enclosure labelled as Canute’s camp. Danbury Morris and Camden (1607) (in No Morris and Buckley state that perhaps Buckley (1978), Sutton 2004), any Viking occupation was aceramic. Medlycott (1993) Morant (1768) (in Morris and Buckley 1978) Pandal Wood Laver (1930) Chalkley Gould No Certain features of the site appear to (1903) correspond to other theorized Viking A Feet of Fines sites. The proximity to a possible for 1206–1207 Roman road may also be considered mentions “Suncastre”, to be influential. believed by Laver to be this site. Benfleet N/A Camden (1607) Yes - ASC A, Chronicle of Proximity of the site to water. A legend (in Sutton 2004), Æthelweard dating from ca.1850 concerns the Spurrell (1885) discovery of skeletons and burnt ships during railway construction, which are thought to be the remains of men killed when the Anglo-Saxons attacked the fort in 893. Shoeburyness N/A Camden (1607) Yes - ASC A The size of the camp may correspond (in Sutton 2004), to the two armies that are supposed Spurrell (1885) to have gathered there. West Mersea N/A N/A Yes - ASC A None Greater London Greenwich N/A Harris (1719) Yes - ASC E None Lysons (1796) Hertfordshire Danesbury N/A Clark (1884) No None Frogmore Lodge, N/A Pollard (1906) No Finds in the local vicinity of “Viking” Aston weapons used to suggest some truth to the claim, but the interpretation of the artifacts is open to discussion. Kent Isle of Sheppey N/A N/A Yes - ASC A, B, C, E, F None Rochester N/A N/A Yes - ASC A, E None Swaines Down N/A Hasted (1782) No None Castle Toll, Davison (1972) Kilburn (1659), Yes - ASC A, E, Chronicle of The site described by Chalkley Newenden Chalkley Gould Æthelweard Gould may be enclosed by a lar ger (1908) earthwork, thought by Davison to represent the Burghal Hidage fort of Eorpeburnan, recorded as being at tacked by the Danes in the Anglo- Saxon Chronicle. Lincolnshire Manwar Ings, N/A Marrat (1816), No None - This is now believed to date Swineshead Thompson (1856), from the 12th century (English Wheeler (1896) Heritage 2012a). Hubbert’s Bridge N/A N/A No None - this is a local legend. 2013 B. Rafffield 29 Site noted in contemporary County/site name Modern source(s) Original source(s) annal/chronicle? Other evidence Gainsborough N/A Godfrey (1666) in Yes – ASC E Allen states that in 1815–1816, Allen (1834), workers uncovered artifacts Stark (1817) including a knife, a battle axe “much resembling an Indian tomahawk” (Allen 1834:26) and a horseshoe. Barton Upon Humber Youngs (1982), N/A No The enclosure partially lies underneath Bryant (1994), a St. Peter’s tenth-century church Reynolds (2003), and Christian cemetery, indicating a Rodwell and date in the 8th–9th centuries. Rodwell Atkins (2011) and Atkins postulate this to be middle Saxon. Norfolk Warham Camp Gray (1933), N/A No A lack of artifactual evidence at the Dyer (1972) site can be interpreted of being a possible indicator of Viking occupation? Dyer states that the regularity of the site is reminiscent of the Trelleborg fortresses. Shropshire Quatford N/A Clark (1884) Yes – ASC A Clark states that adjacent village is named “Danesford”. Suffolk South Cove N/A Morley (1924) No The site has been partially excavated, (possibly the site revealing evidence of a palisade, of Frostenden in medieval sherds and burning. Taken the Domesday Book) by Morley to be a Viking camp and quay. Wales Powys Buttington Musson and Boyd-Dawkins Yes - ASC A, Chroncile of Skeletal remains recovered from Spurgeon (1988), (1873) Æthelweard Buttington churchyard were Smith (2008) interpreted as showing battle trauma. Boyd-Dawkins believed that he could trace the earthwork outlines of the camp. Efforts to locate evidence of the camp and battle have been unsuccessful. Anglesey Llanbedrgoch Redknap (2000, N/A No Human skeletons found in settlement 2007, 2008, 2009) ditch may have been executed—their hands are tied behind their backs. Quantities of hacksilver, weights and other metal objects reveal that the settlement may have been occupied by Vikings. Scotland The Outer Hebrides The Udal, North Uist Crawford and N/A No Small “fort” (7 m across) that seems Switsur (1977), to have been quickly abandoned and Graham-Campbell used for domestic purposes. and Batey (1998)