This article explores the concept of animals, particularly horses in the funerary ritual of the Vikings1 using the example of Pierowall in Orkney, as a means to create and consolidate cultural allegiances in the pioneering stage of Viking expansion. By considering horses in the Pierowall graves not in isolation or solely as inclusions based on religious belief or status, but in the wider social context of the Scandinavian Peninsula and settlement of the Scottish Islands with particular reference to expanding trade networks, alternate interpretations can be made. There are over 130 known Viking pagan burials in Scotland, of which burials containing horse remains (whole or part) comprise approximately seven per cent. The burials containing horse remains appear along the coast of Scotland, often in close proximity such as those on the island of Colonsay in the Inner Hebrides. However, it is only at the cemetery of Pierowall, Westray in Orkney that this burial custom manifests in a significant cluster.

Animals featured heavily in the Viking way of life performing a number of functions and are present in a variety of archaeological contexts including domestic refuse. Animals appear often in the literature concerning Norse mythology, in which a wide variety of species are recognised; the most notable of these animals being the horse. Of the 41 named horses in the Edda, the god Odin’s eight legged horse Sleipnir is the most celebrated. Sleipnir can travel between the worlds, between earth and sky and cross the boundary into the underworld, representing the horse as an active mediator between life and death (Sturluson 2005; Loumand 2006: 133; Jennbert 2011). The significant role played by horses in the burial custom of the Norse is attested to in Gylfaginning where the horse of the god Baldr was burned with his master on his funeral pyre (Sturluson 2005: 25). Ibn Fadlan’s account of a Rus’ chieftain’s (potentially a Scandinavian) cremation details the extraordinary events which take place in the run up to cremation. In addition to the myriad of events taking place including music and drinking, animals are also sacrificed and as part of this ritual two horses are made ‘to gallop until they begin to sweat’ prior to a violent death and being thrown into the ship (Montgomery 2000: 16). In addition to the literature relating to Norse belief, horses are also pronounced in the Viking burial ritual appearing in a variety of grave forms and receiving diverse treatment.

Traditional interpretations of horse remains in Viking burials are seldom questioned. Popular interpretation includes the horse as part of the suite of warrior material culture, that in order to make a fitting entrance into Valhalla the warrior needed not only his weaponry but also his steed (Davidson 1988: 55; Gräslund 1981: 43). The interpretation of horse burials as symbolising personal relationships, wealth, prestige and status has also been advocated (cf. Gräslund 2006; Jennbert 2011). Many of the Æsir of Norse mythology had strong connections with horses, not only Odin but also Freyr (Sturluson 2005), which has led to the interpretation of horses in burial in terms of religious and cultic motivations. The horse in burial has also been considered a mediator in and between all spheres, a means of transport and communication; recently such interpretations have been readdressed making region-specific interpretations (Loumand 2006). It is likely that the horse and horse burial had many connotations but it is not enough to refer only to status and religion – meaning does not necessarily stay the same, it must always be borne in mind that ‘tradition’ evolves and is dynamic, it can be re-interpreted in response to changes and adaptions in society (Leifsson 2012: 190–191).

The use of animals can be a communicative strategy. Animals can be representative and communicative of identity, they can testify to the beliefs, characteristics and status of an individual and community (cf. Pluskowski et al 2010); this is particularly visible in the Viking period in terms of art and personal names for example (Jennbert 2011; Hedeager 2011). Identities are fluid; rather than seeing identity as something people are born with, it is now being considered as an aspect of social relations, something that is learnt, that is adaptable and that can change over time depending on the ways and contexts in which people interact (Jones 1997; 2000; Lucy 2005: 86–87). It is through identity that we perceive ourselves, and how others see us, as belonging to a particular group and not another and being part of a group involves active engagement (Diaz Andreu & Lucy 2005: 2). Animals can also be actors in social relationships, playing an active role in the depiction of identity.

It is a common understanding in funerary archaeology that the dead do not bury themselves, and that ritual enacted at the graveside performs important societal functions. An important strand in this theory is that of the establishment and re-negotiation of alliances and it is through the performance and participation in the funerary ritual that these ends are achieved (cf. Oestigaard & Goldhahn 2006). It has been suggested that burial ritual is one of the first cultural traits to be discarded by immigrant groups arriving into new lands (Laungani 1997). Hadley (2000) and Richards (2001) have discussed the way in which new and indigenous burial rites were adopted by Viking populations and used as social and political propaganda in the Danelaw. Yet in Scotland, there is the suggestion of cultural continuation in terms of burial rite (cf. Redmond 2007), although particular aspects appear to have been ‘cast off’ as is evident in the paucity of animal inclusions in human graves. It is possible that migrating populations might use traditional burial ritual as a means of cultural communication, in creating social networks (Reimers 1999). By consciously using specific burial customs in regionally specific ways, powerful messages could be conveyed to other social groups which may serve social, political and economic ends. Chris Fern has researched the role of the horse in Anglo-Saxon belief and culture and proposes that horse culture and funerary ritual cannot be considered as a simply a ‘passive inheritance’ of earlier tradition, rather it was a custom purposefully re-enacted as a means for the creation of social identity (Fern 2012: 164). When discussing horse cremation in Anglo Saxon England Fern observed an absence of parallels in the period 5^th–6^th century homelands, a time when the ritual was popular in Anglo-Saxon England. He suggests that rather than the rite being a stoical undertaking it can be seen as a response by the Anglo Saxon population to local pressures proposing that such a rite was harnessed when particular conditions became prevalent causing an increase in competition within groups, in this case, in terms of ancestral and political dominance (Fern 2012).

In order to transmit social propaganda, the funerary ritual needs to be embedded within the memory of those participating and those spectating. The concept of performance and mortuary drama is an aid to memory in Viking societies; it is through performance that a preferred identity and a framework for future social relations and interactions can be established. The concept of drama as an element of Norse religion has long been recognised (c.f. Phillpotts 1920) and was revitalised by Terry Gunnell’s The Origins of Drama in Scandinavia (1995). In this and later works (c.f. Gunnell 2006; 2008) Gunnell proposes that drama and performance were a central element in Eddic poetry, that these poems were not only experienced orally but also visually through a cast of actors received by an audience. How such performances were understood depended on the context that shapes it whether this be social, geographical, historical or cultural (Gunnell 2006; 2010). This stance is further developed with reference to other mediums such as festivals, drama and sport. Gunnell also addresses the archaeological evidence for performance citing evidence for animal disguise such as masks and depictions of processions as serving a performative function (Gunnell 1995: 36–80). The concept of drama in mortuary ritual has been explored in Anglo-Saxon burials, particularly at Sutton Hoo, taking place at ‘theatres of death’, a stage for political statements and communicating allegiance (Carver 1998; Price 2008: 156–158). Fern builds on this concept, proposing that burials are active media which are not only statements about wealth and status but can be representative of political intent (Fern 2012: 171). For societies in the past performance was politically, socially and culturally significant with the power to communicate conventional meaning but also to transform it (DeMarrais 2014).

The concept of mortuary drama in Viking age burial ritual has been addressed, principally by Neil Price, who building on Gunnell’s work argues that burial rituals were complex funerary dramas which were acted out at the graveside where stories of memory and constructed history were created and cemented (Price 2010; 2014).

Bertašius (2012) in his discussion of horse burials in Lithuania during the 8^th–12^th centuries AD describes the act as a public ritual in which socially important actions took place and formed the basis for collective memory. Elements of the burial practice would have been familiar to everyone present and would have contributed to the creation of an accumulated and group memory. Bertašius proposes that such a spectacle served to create a shared memory and shared history, cementing social structures and contacts between groups in society and creating bonds (Bertašius 2012). This concept can be applied to the Viking age where the act of horse burial could be considered to consolidate social relations and create memory and stories that would be disseminated through the Viking trading world via the nature of their oral culture and society. Mortuary practices have been increasingly studied in terms of memory. Fern suggests that burials operated as mnemonic events which were suited to creating, reinforcing and disseminating social memory (Fern 2012: 171); the use of animals in this ritual contributed ‘life-blood,’ the animal would be ‘good to remember with’ (Williams 2005: 19; Fern 2012: 172). Horse burial may also have held this mnemonic function at Viking age Pierowall.

Pierowall is situated on the island of Westray, Orkney (see Figures 1 and 2). Westray is one of the largest, northernmost islands in the archipelago. The exact location of the cemetery site, commonly referred to as Pierowall is unknown, but it is certainly located in the area known as The Links, which lies between the grassy covered sand dunes between the modern day village and the site of Quoygrew (Barrett 2012: 26; also see Graham-Campbell & Batey 1998: 130 Figure 7.8). The burials have been dated to the mid-9^th–mid-10^th centuries AD (Sikora 2004: 102). The site was subject to various different antiquarian excavations over the period 1839–1863 (RCAMS 1946: 353–354; Grieg 1940: 5; Graham-Campbell & Batey 1998: 131). The records for these burials were somewhat confused. In 1965 Arne Thorsteinsson presented a paper to the Viking Congress in which he carried out a detailed reassessment of the data and reconstruction of the burials (Thorsteinsson 1968). Thorsteinsson reconstructs a cemetery of 17 graves, however, the cemetery is likely to extend beyond this and contain many more graves. A number of reports testify to the presence of further graves/finds in the vicinity including Reverend James Wallace who wrote that ‘in the Links of Tranabie in Wetra, have been found graves in the sand, (after the sand hath been blown away with the wind) in one of which was seen a man lying with his sword on the one hand, and a Daneish ax on the other, and others have had dogs, and combs and knives buried with them …’ (Wallace 1883: 30). The Reverend George Low also reported graves in the area, noted during the 1700s, as containing the bones of horses and dogs in addition to weapons, beads, brooches and combs (Low 1915: 13) which are likely to be Viking in date.

Of the 17 graves reconstructed by Thorsteinsson, three contain the remains of horses. There is another grave in the area, discovered on the Sand of Gill by Petrie in 1841 which also a contained a horse and is likely associated with the larger cemetery on The Links, totalling a known four graves containing the remains of horses. This is a significant proportion of the known pagan Viking graves of Scotland to contain horse remains, with the likely scenario that there are many more. The majority of the horse burials occupy similar landscapes, not only on this significant trade route, but also in areas which are dominated by wide, sandy bays (e.g. Kiloran Bay, Reay).

Thorsteinsson’s graves numbered seven, eight and 17 contained part/whole horse remains within the grave assemblage. For the majority of the graves orientation is unknown but where known most are aligned South-North (Thorsteinsson 1968: 164–171). Only one grave records the position of the horse, grave number seven. The skeleton of the horse ‘lay quite entire … The horse was laid on its belly, with its head towards the sea, and directed north-east, with its hinder part toward the south-west. The horses head was resting on its nose. The human skeleton was lying immediately before the horses head’ (ibid: 167). In addition to the horse remains the grave also included a bridle bit situated in the jaws of the horse, the remains of a dog, a belt buckle and a spear head (ibid: 167–168; Graham-Campbell & Batey 1998: 133). Grave eight contained part of a human skeleton with part of a skeleton of a horse; no other information is provided on the burial other than the presence of a bridle bit and knife (Thorsteinsson 1968: 167–168; Graham-Campbell & Batey 1998: 133). Grave 17 is classed as a boat burial, with 21 rivets found during excavation; parts of a human skeleton with that of a horse were discovered with other items including a bone button and horse fittings (Thorsteinsson 1968: 171; Graham-Campbell & Batey 1998: 134). Thorsteinsson’s reconstruction of the graves does not provide any detailed information for what he identifies as grave number five, but is reasonable to assume that this is the grave excavated by George Petrie in 1841 ‘on the sand of Gill’ (see above) which contained the skeletons of a man and horse and also contained a shield boss (RCAMS 1946: 353).

Although little information is provided in the accounts as to grave form it had commonly been assumed that the graves were under mounds but Thorsteinsson’s reconstruction suggests that few had been covered (Thorsteinsson 1968: 163). Stone packing and square stone settings are noted in some of the Pierowall burials (ibid).

Pierowall was located on the main sea route from Scandinavia through to the Western Isles of Scotland into the Irish Sea region (Owen 1999: 27; Wilson 2008: 109). This was an important route for trading, raiding, and military expansion. This route, like others, is a voyage which is taken regularly, underpinned by the knowledge and experience of travellers which had been embedded through previous journeys and verbal exchanges (Sindbæk 2012: 153).

Orkney reveals little evidence for the existence of market places although the bay at Pierowall has been proposed as a beach market (Owen 1999: 23) (See Fig. 3). Similar locations along the coast of Britain and Ireland have been interpreted as maritime havens, economically and politically strategic locations which may have functioned as raiding bases, trading stations, havens for repair and shelter or a combination of such purposes (Sheehan et al. 2001). The bay at Pierowall provided an ideal location as a strong and secure base between the homelands and the Western Seaboard, providing a good post from which to mount raids, for trading, and as a convenient base for political dominance in the establishment of the Orkney earldom (Brøgger 1929: 121; Crawford 2013: 89; Dalland & Owen 1999: 177; Grieg 1940: 6; Owen 1999: 26). This bay is also the best natural harbour in the northern isles of Orkney (Owen 1999: 23). The village at Pierowall was known at the time the Orkneyinga Saga was written down (Palsson & Edwards 1981: 133), at that time known as Hǫfn (Old Norse ‘harbour’) (Grieg 1940: 6; Barrett 2012: 26). Further to the graves detailed above, there are also finds from Pierowall which are evidence of interaction with the maritime traffic through this important sea route; one example is an Irish style brooch of which similar examples have been found in Ireland, Wales and Norway (Graham-Campbell & Batey 1998; 134).

Figure 1 depicts the distribution of pagan Viking graves across Scotland which were reported to contain horse remains. In addition to the four burials at Pierowall, horse remains were also found in the following graves: at Ballindalloch (Viking date uncertain) (Anderson 1874, 569; Graham-Campbell & Batey 1998: 105); at Reay, Caithness, the ankle bone of a horse was discovered in addition to other bones in the sand although it is unclear which burial the bones were associated with (Edwards 1927: 202; Graham-Campbell & Batey 1998: 125); on the islands of Vatersay and Tiree horse remains were reported but no formal site record exists (Graham-Campbell & Batey 1998: 83 & 87); on the island of Colonsay there are three graves containing horse remains, at Machrins, Cnoc Nan Gall and Kiloran Bay (Anderson 1907: 447; M’Neill 1892: 62; Ritchie 1981: 278).

The burial at Kiloran Bay was that of a male aged c. 40 years who had received the rite of a wealthy boat burial whose grave goods amongst others included Anglo-Saxon stycas, a silver cloak-pin of Norwegian type, some knives including one known to have a pivoting blade of Anglo-Saxon or Anglo-Scandinavian type, scales and lead weights most of which have decorative mounts including fragments of Anglo-Saxon and Irish ornamental metalwork, bronze harness mounts of Insular type and a shield boss of Irish Sea type (Graham-Campbell & Batey 1998: 119–120). This selection of grave goods shows considerable contact outside the Scandinavian lands and Scotland. The burial has been interpreted at that of a trader or warrior (ibid: 122), indeed such a wide variety of items may have been accumulated by a trader travelling across the North Atlantic and navigating the western seaboard, furthermore the presence of scales and weights may be testament to his role. The burial has been dated to the end of the 9^th century (ibid). The burial architecture also included a chamber within the boat (Anderson 1907: 443; Bill 2005: 348; Graham-Campbell & Batey 1998: 118–119) reminiscent of the chamber graves at Birka, also noted at grave ten at Pierowall. Also included in the grave assemblage was a horse, a healthy animal aged between six and eight years old and estimated to be approximately 14.2 hands high (c.1.46m) (Graham-Campbell & Batey 1998: 120).

Following on from earlier discussions of burial as performance and of memory to make political and social statements of identity and communicate allegiance, can we decipher alternative interpretations beyond wealth, status and religious associations for the practice of horse burial in Viking Scotland? To answer this, it is necessary to consider the wider context, specifically the wider Viking world considering the social, political and economic climate of the period.

The period contemporary with the cemetery at Pierowall (c. AD 850–950) was the early settlement phase in Scotland; it would have been a time of considerable upheaval, both socially and politically. There was a native population present in the lands settled by the Vikings in Scotland meaning interaction, whether on peaceful or forceful terms, was in progress. It was a period in which raiding was prevalent and trade was expanding; these acts were not mutually exclusive occupations but interdependent activities (Griffiths 2010: 30 & 100; Sindbæk 2012: 150). When a ship was appearing in the bay you did not know whether their intention was to raid or trade. Viking allegiance was by nature very fluid. The Viking armies and their campaigns across Europe are evidence of this; armies were multi-ethnic, they included women and families, they were not single entities and had no sense of national solidarity, the armies were in a constant process of reform and allegiance could switch (N. Price & B. Raffield 2014, pers. comm.). Due to the fluid nature of allegiance it would have been important to signal shared cultural norms to the social groups passing through this busy maritime highway.

Trade was extremely important in Viking Society and had multiple functions which were political, economic, and social. The systems of communication that connected trading places were complex. Trade took place along established routes between specific locations, between main trade centres and smaller sites. The structure, geographically, of the trade networks ultimately lay in the hands of individuals; those involved in long-distance trade had considerable motive to seek what was considered the most suitable, safe and active places for trading; this in effect would mean that those involved in long-distance trade would seek the same few sites (Sindbæk 2012: 154; Sindbæk 2007: 128). Sindbæk proposes that trade networks were bound by personal and local ties. Such networks were based on relationships of trust facilitated by symbolic communication creating and demonstrating shared cultural norms (Sindbæk 2012: 154–155). This is also supported by the operation of a shared bullion economy where weights used in this system are also found in graves including Scar on the neighbouring island of Sanday, Orkney (Dalland & Owen 1999: 118–126). Those participating in raiding and trading excursions would have required experience, those who knew the routes, safe harbourages and points of exchange. This concept of symbolic communication and shared cultural norms could be extended to burial ritual, through the use of the horse as a common, unifying symbol.

Scotland at present yields no evidence of the large urban centres involved in trade and production visible in other areas of the Viking world such as Kaupang and Birka. Trade occurred at smaller market places. The concept of the beach market was a familiar one in the Viking age, particularly in the Irish Sea region where two important beach markets were located: Meols on the River Dee and Whithorn in Galloway, possibly functioning as specialised market sites (Griffiths 2010: 110–118; Wilson 2008: 116). Small markets such as these would have been in competition with each other (Wilson 2008: 53) and accordingly this would have necessitated the need to identify as the favourable trading partner. It would therefore be in the interests of the community at Pierowall to attempt to secure trading links and to identify with a particular social group, creating strong allegiances which were cemented by similar practices and beliefs which would have been particularly familiar to those involved in long-distance trade from centres such as Birka in Sweden and Kaupang in Norway.

The town of Kaupang was established c. AD 800–960/80 (Skre 2007: 18). This trading town was situated in a protected bay on the main sea route from the coastal areas inland. There are a total of eight cemeteries surrounding Kaupang which exhibit wide variation in burial rite including boat burial, a possible chamber burial, cremations and inhumations in storage chests, toboggans and log coffins (Stylegar 2007: 65), spanning a period of c. AD 800–950 with a small preponderance of 10^th century graves (ibid: 81–86). The total number of graves is estimated at some 700, although this is likely to be an underestimate (ibid: 77). The alignments of the graves are only known from the Bikjholberget cemetery with the overwhelming majority aligned NNE-SSW or N-S (ibid: 88). Flat graves and boat graves are relatively rare in the region, yet at Kaupang the number of graves of this form is striking and particularly prevalent at the cemetery at Bikjholberget (Stylegar 2007: 103–128; Skre 2012: 114). Stone packing of burials is also found at Bikjholberget (Stylegar 2007: 88) and also present in the Pierowall cemetery. It was also this cemetery (which contains many unexcavated graves) that contained the majority of burials containing horse/horse remains and a wealth of imported artefacts (ibid 2007: 103–128; Skre 2012:114). The nature of trading networks and expansion is also indicated by the imported grave goods found in the Kaupang cemeteries: in the graves dated to the 9^th century items from the continent were predominant, whereas the graves from the 10^th century saw a shift with items of Insular and eastern production gaining equal important to those of continental manufacture (Stylegar 2007: 65).

The trading town of Birka, centred on a small island in the bay of the Baltic Sea, now known as Lake Mälaren in eastern central Sweden, was established in the 8^th century AD and abandoned at the end of the 10^th century (Wigh 2001: 20). As at Kaupang, there were a number of cemeteries associated with the town of Birka. The cemeteries at Birka depict wide variation in burial rite including cremation and the more unusual practice of inhumation in wooden coffins and in chamber graves (Ambrosiani 2012: 97; Gräslund 1981). In 1981 Gräslund estimated the number of graves on Birka to total some 2,300 (Gräslund 1981: 4). This total is now gauged at over 3,000 known graves (Wigh 2001: 17). The burials assessed by Gräslund were broadly dated based on graves goods to the Early period at Birka and the Late period at Birka, covering the early 8^th–mid 9^th century and late 9^th–10^th century respectively (Gräslund 1981: 3). Gräslund (1981) reported that in two of the seven cemeteries, one North of Borg and another at Hemlanden, a total of 20 graves which included horse remains were excavated, the majority of which lie at Hemlanden and all of which are chamber graves. Horse was the only animal interred in the inhumation graves at Birka (although other animal species are found in cremation deposits). The chamber graves are most common in the late period at Birka; all chamber graves which contain horses which are dated are also dated to this later phase (ibid.) and contemporary with the burials at Pierowall. Nearly all the horses in the Birka cemetery are placed on their left side with head pointing south and turned towards the chamber, but there are some which differ in orientation and position, most notably graves Bj834 and Bj581; both horses in these graves were laid on their bellies with their legs folded (ibid: 41). This unusual placement of the horses is mirrored in the Pierowall assemblage. It was suggested that different groups of people were buried in different cemeteries around Birka; in the case of the chamber graves, these may be prominent merchants (Gräslund & Müller-Wille 1992: 187).

Orkney would have provided an ideal base for raiding and initial settlement as the Vikings began to colonise Scotland and the North Atlantic. As the Viking world expanded particular places began to operate as specialised market places and bases for trade. With the expansion would have come increased competition with markets striving to be the preferred market on what was an expansive seaway, with numerous potential locations for competing market places. A trader would seek the most favourable, safe and active place for trading. Such networks were partly based on trust, which was facilitated by symbolic communication. Burial ritual was an extremely effective method of communication, with horse burial taking on an essential role within the burial performance, giving ‘life-blood’ to the event, serving as an aid to memory which would have been transmitted through narratives and the oral culture of the Vikings. The use of horses in the burial ritual would have been a familiar act with merchants travelling along this main sea route and from the main trading centres. Horse burial and the associated ritual served to create a collective memory that linked people, and it was through this practice that future trading links could be secured.

The use of animals in a political context, utilised to create and re-negotiate alliances, is not unknown in the Viking world. Horse fights are cited as one of the many Viking age pastimes (Gardela 2012: 242) and in the saga literature feature as dramatic episodes which perform a number of functions within social dynamics, one of which had political ramifications: while functioning as loci for violent conflict, the humiliation of one’s enemies at horse fights could prove an effective political strategy which could attract new allies and serve to strengthen existing alliances (Martin 2003: 28–40). A further example comes from the site of Hofstađir, Iceland, where excavations revealed 23 cattle skulls. Osteological analysis of the remains indicated the cattle were subject to an unnecessary, violent and dramatic death with the skulls then placed on display. Lucas and McGovern (2007) argue that this violent and bloody decapitation served a political function by dissipating conflict and social tension, and through the seasonally repeated slaughter and display of the remains, a long-term memory was created which added to the history and role of the site.

The cemetery at Pierowall, despite the nature of the record, exhibits a number of similarities with traders and trading communities, particularly those of the Kaupang and Birka. The burials are Pierowall are broadly contemporary with the trading centres and associated burials at Kaupang in Norway and Birka in Sweden. This was a period of expansion of the important trading route through Orkney to the western seaboard of Scotland, a move further supported by an increase in imported items and items of Insular manufacture in the graves of these central trading hubs. They also depict a number of similarities in burial rite in terms of horse burial, but also marked differences which leads one to believe that those interred at Pierowall are not those hailing from these centres (for we would expect greater homogeneity) but emulating elements of their practice. This is further corroborated when we consider other burials which include horse in the assemblage along the coast of Scotland and western seaboard to Ireland, particularly the warrior merchant of Kiloran Bay. Both the cemeteries surrounding Birka and Kaupang contain horses in burials and it has also been suggested that these animals played an essential role in the funerary drama of such places. For the single graves for which the position of the horse is recorded at Pierowall, we have examples of the same unusual placement at Birka. It has been proposed that the burials at Birka featuring horses were those of prominent merchants, indicating trading links. The burials at Pierowall demonstrate a similar grave form to Kaupang, and in particular Bikjholberget, in which flat graves are the dominant form, where stone packing is also common. The inclusion of a horse boat burial is also common at Kaupang where the majority of horse inclusions are in boats.

We can perhaps also see the Pierowall burials as mirroring that of traders travelling through this well-established trade route. The burial and assemblage at Kiloran Bay indicates the individual was a trader, perhaps on a voyage through the seaway. With his trading equipment he was also interred with a horse, on the favoured wide sandy bay, reminiscent of trading centres in the Scandinavian Peninsula. If the man buried in this grave was a trader, on a voyage through this trade route, he was perhaps the type of merchant which the community at Pierowall were seeking to emulate.

This article illustrates the potential and alternative interpretations that can be made from horse remains in burials in Viking Scotland. It is likely that horse burial had many connotations therefore this paper does not seek to dismiss traditional interpretations. Rather it highlights that by considering the act of human-horse burial in the wider social context and not in isolation within the grave, interpretations beyond status, wealth and religion can be deciphered. Perhaps we can interpret the act of horse burial at Pierowall as a powerful politically, economically, and socially charged event, depicting a particular form of identity transcending the traditional interpretations of horse burial, such as status and as a warrior, to convey a message of allegiance. By considering Pierowall in the wider social and political context, and making comparison to principal trading places we can perhaps interpret horse burial at Pierowall as serving a communicative function; portraying the population and community at Pierowall as suitable and preferable trading partners.