Insular artefacts from Viking-Age burials from mid-Norway . A review of contact between Trøndelag and Britain and Ireland

This article presents a detailed overview of the Insular artefacts found in Viking-Age burials from the Trøndelag region of mid-Norway, most of which have not previously been published in English. The archaeological evidence indicates that contact between Trøndelag and the British Isles was well established at an early stage of the Viking Age. The main evidence for contact comes from the 9th century, when a number of significant patterns can be discerned. Some local concentrations of Insular goods show the continuing importance of some preViking centres, while other areas suggest co-operation between several neighbouring families in order to equip and provision overseas expeditions. Later, the datable Insular artefacts indicate significant changes in the nature of contact. North Sea trading towards the end of the Viking Age appears to be affected by increasing centralisation of power in Trøndelag during the 10th century.


Summary
This article presents a detailed overview of the Insular artefacts found in Viking-Age burials from the Trøndelag region of mid-Norway, most of which have not previously been published in English. The archaeological evidence indicates that contact between Trøndelag and the British Isles was well established at an early stage of the Viking Age. The main evidence for contact comes from the 9th century, when a number of significant patterns can be discerned. Some local concentrations of Insular goods show the continuing importance of some pre-Viking centres, while other areas suggest co-operation between several neighbouring families in order to equip and provision overseas expeditions. Later, the datable Insular artefacts indicate significant changes in the nature of contact. North Sea trading towards the end of the Viking Age appears to be affected by increasing centralisation of power in Trøndelag during the 10th century.

Introduction
The Viking Age (c. AD 800-1050) is commonly associated with the extensive, if not always peaceful, Scandinavian interaction with the British Isles. The Norse influence and involvement around the Irish Sea is well attested through contemporary literary sources, as well as from a varied archaeological record. In Norway, the main body of evidence for this contact comes from the many Norse burials that contain a wide range of Insular artefacts. So far, over two hundred such burials have been identified, but the number of individual pieces is far greater (Wamers 1985). This article seeks to present an overview of Insular grave goods found in the Trøndelag region of central Norway, most of which have not previously been published in English, and to discuss the nature of contacts reflected through this material. Despite the quantity of material previously recognised, very little research has been done on the subject for this region. Traditionally, researchers have focused their attention on the Insular artefacts from the trading settlement of Kaupang near Oslo and from graves in Western Norway, while the Trøndelag area, with notable exceptions, has been largely overlooked. Hence, this is the first study that has examined the whole group of Insular finds from this region, and provides essential primary data on the range of artefact types and their distribution. One of the purposes of this article is therefore to draw attention to the significant and under-reported corpus of Insular metalwork. The increased data also gives a new and stronger basis to review contacts with Britain and Ireland, and to suggest how these developed over the period.

The Trøndelag region
Trøndelag as a region is dominated by the Trondheimsfjord, an area rich in agricultural land and in a sheltered location, with excellent communications by both water and land. The fjord itself would have made the coastline easily accessible for communities based in the inner parts of Trøndelag, while several large valleys connect the fjord with central Sweden. The location of numerous monumental burial mounds, affluent graves, cult and court sites within the region, suggests that the Trondheimsfjord was an area of complex economic and social development in the Viking period, with well-established contacts both westward and to the east. During the high and later medieval periods, Trøndelag was significantly smaller than it is today, and the region was divided into eight fylkir (counties), comprising Inntrøndelag and Uttrøndelag (Figure 1). The northern area of modern Trøndelag was part of the county of Namdalen, while Fosen, the coastal area of southern Trøndelag, formed an element of Nordmøre county. However, this division is likely to have been formed well before the high medieval period, suggesting a well-established and organised power structure in the late Viking Age (Skjevik 1997, 185-7).

The Insular Finds from Trøndelag -An Overview
Many of the finest pieces of Insular Viking-Age metalwork have been found buried in Norwegian ground, as the pagan funeral custom of furnished inhumation or cremation continued until the middle of the 10th century at least. An updated catalogue of the Insular material in Trøndelag shows that there are 86 Insular objects at 68 confirmed or likely burial sites, most of which lie close to the Trondheimsfjord (Figure 2, Table 1). The descriptions and results presented in this article stem from the author's graduate research completed at the University of Trondheim in 2013 (Heen-Pettersen 2013). The Insular material is described by broad artefact type in Table 2, highlighting notable concentrations, while some previous and current theories are presented concerning their use and meaning in contemporary Norse societies. Individual site names are given in the format of farm followed by township, for example Skei, Steinkjer.  A wide variety of grave types contained such material, including barrows, boat burials, cairns, coffined burials and cists, in addition to simple dug graves. The majority of the Insular artefacts have, however, been found in barrows, many as a result of the numerous excavations of this burial type during the latter half of the 19th century. Most of the Insular artefacts from mid-Norway excavated during this period were purchased by the archaeological museums of Oslo and Trondheim, but some finds were sold abroad. A number of artefacts from Viking-Age burials in Trøndelag are therefore owned by museums outside Norway, including the British Museum. However, as a consequence of the discovery of the complete Viking ship burial at Oseberg in 1904, the law was changed the following year to protect all archaeological finds. The farmer who owned the land on which the Oseberg burial was discovered wanted to sell the finds to the highest bidder, and, according to rumour at the time, the potential buyer was said to be English (Fagerland 2008, 297). With the new law enforced, archaeological remains dating from before AD 1537 were given automatic protection and would now formally be owned by the Norwegian state. Insular artefacts found in Trøndelag after 1905 have therefore been officially administered, recorded and stored with the Archaeological Museum in Trondheim, today called Vitenskapsmuseet (Museum of Natural History and Archaeology).

Decorated bronze/silver mounts
The largest group of Insular objects found in Norway, and also in Trøndelag, are various types of highly decorated pieces of mounts. In Trøndelag, 28 whole or partial mounts have been found, widely spread across the region. Most of these items appear to have been part of horse harnesses or from religious equipment, such as reliquary shrines, books and altar furniture (Figures 3-6). All have traces of gilding, and, with the exception of two silver examples, all the pieces are made of copper alloy. The majority of the mounts are thought to be of Irish origin (Bakka 1963;Wamers 1985). Egil Bakka and Egon Wamers have, however, identified four pieces from Trøndelag that may be English products ( Figure 35 and Figure 36). Three of these have been found in the region of Stjørdal, an area with an unusually high concentration of English-produced objects (discussed in section 4.2). A great many of the decorated mounts appear to have been converted into brooches by the addition of pin fittings. However, there are also three examples from Trøndelag -from Finsås in Steinkjer, Grande and Uthaug, Ørland -where the mounts were buried at the waist of the women, accompanied by keys and knives, indicating their use as belt decorations (Heen-Pettersen 2013, 51-53). With few exceptions, Insular mounts are found in 9th-century women's graves. Only four male burials from Trøndelag are known to have contained such objects. Traditionally, such artefacts are often interpreted as exotic gifts or souvenirs given to the wives and mothers of men returning from raids overseas. According to Johannes Bøe (1927, 15) 'The custom of wearing souvenirs presented to them by the Vikings on the return from his westward expedition, has evidently been high fashion among the Norwegian women in the first half of the 9th century'. However, a number of Insular mounts have been found in graves from the contemporary trading settlements of Kaupang, Hedeby, Helgø and Birka, leading to the suggestion that some of this material represents merchandise and was sold second-hand (Blindheim 1978;1999). Alternatively, this material may also derive from the Norse custom of 'pillage-trading', which means that people who traded also took part in raids wherever this was more profitable. Wamers (1998, 44) has therefore proposed that the presence of Insular 'loot' on trading settlements could indicate the presence of such traders turned raiders. In any case, based on the dates of the graves containing pieces of mount, it seems that these objects were most valued and appreciated during the first phase of contact and colonisation of the British Isles. Such pieces worn as jewellery must indeed have appeared as exclusive and exotic forms of dress ornament, with a clear association to the Insular world. Whether used as brooches, belt-fittings or other types of decoration, these items would have formed a prominent part of the clothing they were worn on. Various types of dress ornaments are often highlighted as artefacts likely to have been utilised as distinctive and important symbols of the wearer's status and social background (e.g. Gustin 2004;Harlow 2004). This may have especially been the case for exotic items, which were obtained from remote landscapes (Hines 1994;Glørstad 2010). In that connection, it is worth noting that three of the four earliest datable graves from Trøndelag with evidence for contact with Britain and Ireland, each contained pieces of mounts. The converting of such Insular objects into ornaments of Norse dress at the very beginning of contact emphasises the strong meaning these foreign artefacts had in the early Viking Age.

Ring brooches and ringed pins
The desire to wear Insular dress ornaments is also reflected in the small number of ring brooches and ringed pins found in Trøndelag. Dress fasteners in the form of bronze ringheaded pins represent a widespread male fashion and were used to fasten an outer garment, e.g. a cloak. The commonest variety in Norway was the plain, ringed loop-headed pin, and it has its origins in Ireland in the pre-Viking period (Fanning 1994, 15-23). Ringed pins found in Norway are often regarded as Insular imports (e.g. Jåtten 2006;Geber 1991;Grønnesby and Ellingsen 2012). However, their simplicity complicates the problem of distinguishing imports into Scandinavia from their local copies (Graham-Campbell 1984, 35-38). Of the ringed pins from Trøndelag, three have rings with cubic knobs terminating in points and were regarded by Petersen (1940, 203-4) as probable Insular imports (Figure 7). The origins of four simple pins are more problematic. However, it can be noted that none of the simple types from Trøndelag display certain characteristics associated with Scandinavian copies, like the combination of large ring with a short pin or decoration in the form of rows of ring-and-dot stamping down its short pin (Graham-Campbell 2011, 103). The simple ringed pins from Trøndelag are therefore considered here as possible Insular imports, although this supposition must naturally be regarded with caution. In that connection, it can be noted that one of the simple ringed pins from the region was recovered together with an Insular balance scale from Lø, Steinkjer, perhaps suggesting a foreign origin for the pin as well.
Only 22 examples of imported Insular ring brooches have been identified in Norway, and four of these are found in Trøndelag (Glørstad 2010). One example comes from the coastal area at Nes, Bjugn, while the remaining brooches, from Snåsa, Fossem and Smolan, are concentrated in the inner part of the Trondheim fjord, within the old area of Inntrøndelag. The Norwegian term 'ring brooch' is used for convenience here, but recovered examples include both annular and penannular forms showing considerable variation. This group of artefacts is almost exclusively found in women's graves dated to the 9th century. In some cases, the Insular ring brooches have been found in situ on the middle of the chest or under the chin of the skeleton,  (Glørstad 2010, 210). The traditional dress of a wealthy Scandinavian woman during the Viking Age would have required the use of two substantial oval brooches to fasten the shoulder straps of a tunic worn over a shift, while a woman's third brooch -in the above cases Insular ring broocheswas used to fasten a cape or cloak. James Graham-Campbell (2001, 36) has suggested that the incorporation of Pictish and Irish broochtypes as cloak-fasteners in Norse dress might be viewed as 'glittering prizes'the use of the exotic to make statements concerning status. Ann Zanette Glørstad (2010, 124) has, however, put forward an alternative explanation, suggesting that Insular ring brooches might have been brought to Norway at a time when the journeys to the British Isles had become more routine and where women themselves participated. Following this interpretation, the women buried with Insular ring brooches might have been Norse women communicating their personal connection with overseas colonies through their jewellery. Such items would display the economic and political resources that could be drawn from these Insular networks (Glørstad 2010, 38). In this connection, it is worth noting that two of the four known ring brooches from Trøndelag -found at Nes, Bjugn and Snåsa -represent the finest and best-preserved examples of this brooch-type found in the whole of Norway (Figure 8, Figure 9). The prestige of owning, and being buried with, such items may well indicate that the women, or at least their families, had a particularly important role within these overseas networks.

Weighing equipment
From Trøndelag, 12 burials are known containing remains of balance scales and weights. Balance scales most commonly occur in men's graves, of both the 9th and 10th centuries. In Trøndelag nine examples are known from male burials, while only two such items have been recovered from women's graves (Heen-Pettersen 2013, 56-57). Burial of unburnt bodies seem to have been the prevailing custom in Trøndelag for those with Insular weighing equipment, and in at least six cases, the body was placed in a boat covered with a mound. Several of these locations stand out because of their particularly rich grave-furniture, emphasising the buried person's importance within their families and communities (Herstad 2012). Associated grave goods include various types of weapons, e.g. high-quality swords with silver decorated handles; two examples from Haug, Verdal, and Gravråk, Melhus, which have Ulfberht inscriptions on the blade; personal jewellery of silver, glass, amber and carnelian, in addition to various types of tools and equipment, e.g. whetstones, blacksmiths' equipment and sickles (Herstad 2012, 101).
Several researchers have claimed that most of the scales found in Norway may be of Insular origin, based on their decoration and design (e.g. Jondell 1974;Petersen 1940). The fact that several of the graves also contained other artefacts of Insular origin may support this assumption. Sometimes combined with bronze boxes and lead weights, the scales have often been associated with traders and generally interpreted as evidence of Norse trading activities in the western world. However, it has also been suggested that such weighing equipment may have been necessary for accurate loot-sharing (Wamers 1998, 43). Their small size suggests the weighing of valuable commodities. During the Viking period, this might have principally been silver, whether coins, ingots or hacksilver (Gaimster 1991). Within the Trondheimsfjord area, balance scales are found at nearly every place where marked concentrations of Insular material occur. It is, however, noticeable that in most cases only one burial within each of these areas contained such weighing equipment. This suggests that trading, or other activities involving the weighing of small, valuable products, was concentrated in only a few persons or families within each community. A particular well-preserved balance scale from Trøndelag was found in 2003 during an excavation at Lø, Steinkjer ( Figure 10). Here a richly furnished male inhumation of AD c. 900-950 was discovered, placed within a 9.5m long boat that had been dug into the top of an older barrow (Grønnesby and Ellingsen 2012, 21). This is in fact the most recent grave containing Insular artefacts to be discovered in mid-Norway. The grave goods in the Lø burial included a complete set of weapons, a ring-headed pin, some tools and personal items, in addition to the balance scale of tin-coated bronze, which had been put in a small wooden box before being placed by the feet of the man ( Figure 11). The set of scales comprises two bowls, each of which is suspended on three wire chains, which are drawn together towards the point where they are attached to a spacer. Above each spacer there is a further length of chain that joins with the balance bar of the scale. Each bowl measures 65mm in diameter and is decorated with delicately engraved geometrical designs and a rosette ornament on the inside of the base. The decoration has penetrated through the shiny tin coating to reveal the contrasting yellow copper alloy underneath (Skinner 2012, 120-1). Based on the tin-coated bronze, the design and decorations of the arms, chain and scales, Geir Grønnesby and Ellen Ellingsen (2012, 34-35) have classed the scale from Lø as a Steuer type 2 and believe that this particularly example is likely to be of Insular origin. This type of scale is generally placed in the 10th century, and they are only found in Norway, Ireland and Scotland (Steuer 1997, 237).
Weights are not uncommon finds from the Viking Age in Norway, and are sometimes found with accompanying scales, but from Sør-Trøndelag, there are two examples from Tønnøl, Bjugn, and Solstad, Skaun, of particular note. They are made of small, Insular decorative mounts, which have been turned into weights. The weights appear somewhat different in shape, size and weight. The example from Solstad, which represents a stray find, measures 43mm in diameter with a weight of 182.4g ( Figure 12). It had, however, slightly disintegrated when found and the original weight is therefore uncertain (according to finds information in the museum catalogue). The weight from Tønnøl has a conical shape and appears complete, but weighs only 24.2g. It was found together with another lead weight, both of which had been placed inside the bowl of a balance scale when discovered in a male boat-burial, dated to the 10th century (according to finds information in the museum catalogue). In addition to the examples from Trøndelag, such weights are known from at least eight other sites in Norway, including a female burial at Hopperstad in western Norway, which is one of the richest female Viking graves from western Norway (Sørheim 2011, 17). These weights, which originally formed part of prestigious church objects, ended up in Viking society where their function was somewhat different from their original purpose.

Drinking horn vessels
Drinking horns, for serving and consuming alcoholic drinks, are thought to have been an important piece of household equipment in societies where feasting and formal entertainment played a major role (see e.g. Etting 2013). The 7th-century heroic British poem Gododdin refers frequently to feasting and to the use of drinking horns for mead and wine (Youngs 1989, 62). Several of the sagas in Heimskringla (e.g. the saga of Harald Harfagre, Hakon The Good and Olaf Trygvasson) include vivid accounts of royal banquets where the use of drinking horns is mentioned (Etting 2013, 29). In Irish folklore, the possession and use of such items were the preserve of the elite, where status could be demonstrated by the ownership of a large number of horns, of horns displaying certain types or ornamentation, or even of famous named horns (Sørheim 2011, 21). A primary function of these prestige items was to 'enable their owners to demonstrate status by providing unlimited hospitality, an echo of the hospitality obligations mandated by the laws' (De Vagner 1995, 81-82). Finds of Insular drinking horns are, however, sparse and fragmentary; such items have mainly been found in Viking-Age graves in Norway and in unstratified contexts near settlements in Ireland (Sørheim 2011, 21). In addition, they are occasionally found in Norse burials in Ireland and Scotland (Redmond 2007, appendix 5). In Norway, the remains of 24 Insular drinking horn vessels are known, and seven of these come from Trøndelag (Wamers 1985, plus recent finds). Three of the vessels are found in the old county of Namdalen, while the remaining examples are located within the Trondheimsfjord. The horn itself rarely remains, and the surviving evidence consisting mainly of the copper-alloy terminals and mount fittings and occasionally of suspension chains. Insular drinking horns have been found primarily in women's graves dated to the 9th century (Petersen 1940, 11). This is also the situation in Trøndelag, where only one example from Voll, Overhalla, comes from a man's grave. Drinking horns with zoomorphic terminals might have been specially valued for displaying status, and in Trøndelag five of the seven recovered examples display this type of ornamentation (De Vagner 1995, 83). One such terminal of particularly high quality has been found at Varøy, Naerøy ( Figure 13). Also worth noting is the discovery of two drinking horn terminals from a cremation burial at Gjeite, Levanger, since this is one of only two graves in Norway where more than one such item has been recovered. The grave goods from Gjeite also included a large bronze cauldron of Insular origin. A similar combination of artefacts is known from a Viking-Age burial at Birka, Sweden, where two drinking horns and a glass vessel were placed inside a bronze cauldron. It has been suggested that these artefacts represent a drinking set (Trotzig 1984, 227-8), and it is possible that the Insular grave goods from Gjeite had the same function.

Bronze vessels and decorated wooden buckets
In Trøndelag, six Insular bronze vessels -including bowls, dishes and cauldrons -have been found (Heen-Pettersen 2013, 59). This is, however, a relatively small number in comparison with the western part of Norway where more than 30 bronze vessels are known (Jåtten 2006).
A large hanging bowl of Irish origin, found in 1986 at Skei near Steinkjer, represents a particularly rare type, remarkable both for its triangular shape and its massive size ( Figure  14). Its only close parallel is represented by the far smaller triangular copper-alloy hanging bowl from Kilgulbin, Co. Kerry (Graham-Campbell 2001, 29). The Skei bowl's sides have an average length of 410mm and a depth of 130mm. Inside the bowl, a strainer plate is attached, decorated with a pattern of perforations in the form of a central rosette between a pair of roundels. The interior of the bowl, which has been patched, is decorated with ridges, whereas the exterior is mounted with three bird-shaped escutcheons, each having tinned wings and a three-feathered tail ( Figure 15, Figure 16). This hanging bowl was found in a richly furnished woman's burial placed in a stone cist and is dated to around AD 775-800 (Stenvik 2001, 40). It was accompanied by a number of other items, including three other pieces of Insular metalwork: a gilded bronze mount, a large bronze ladle ( Figure 18) and a small bucket of yew covered in decorated bronze plates of the highest quality workmanship ( Figure 19). The Skei bucket is a close parallel to both the well-known bucket from Birka grave 507 (Youngs 1989, no. 120) and a bucket from Hopperstad, Stavanger. Graham-Campbell (2001, 30-31) and Stenvik (2001, 30) have argued that the buckets from Hopperstad, Birka and Skei are so similarin shape, handle mounts and, in particular, the shared depiction of animals and bird in plant-scrolls on the bronze sheets, that they may have come from the same workshop. The bowl, ladle and bucket were found together at the foot end of the Skei grave, with the ladle partly overlying the bucket (Stenvik 2001, 12). The combination of such Insular items also occurred in the remains of a destroyed burial mound at Steinvik, Bjugn in 1894. According to an unpublished letter from the landowner (dated 25 May 1894), the Insular finds, comprising two bronze ladles, a bronze cauldron and fragmentary remains of a yew bucket, were found together under a large stone at the site of the mound, which had been levelled a few years earlier. In addition to the two examples from Trøndelag, the combination of such Insular items is also known only from the previously mentioned woman's burial at Hopperstad, Stavanger. Bødal (1998) has suggested that such artefacts may represent a fine and exclusive set of drinking or serving equipment. Graham-Campbell (2001, 33) has indicated a similar function for the Insular serving equipment from Skei, which may have been used 'on the high table'. Only 12 bronze ladles, probably all produced in Ireland, have been found in Norway and five of these are from Trøndelag, from both male and female burials (Petersen 1940, plus recent finds). Only three of these can be given a more precise date, all of which can be placed in the first half of the 9th century. Two of the ladles were recovered from the grave in Steinvik, Bjugn, while the remaining three examples all come from the Steinkjer area. Raghnall O'Floinn (1989, 122) divides the Insular bronze ladles into two groups: those with long handles and small bowls and those with short handles and relatively large bowls. Only one of the bronze ladles from Steinvik, Bjugn, might be classed as long-handled with a small bowl (Figure 17), while the remaining examples from Trøndelag all belong to the latter group. One detail worth noting is the remains of a runic inscription and two overlapping ship symbols engraved on the back of the handle of the Skei ladle. The runes are believed to represent the three first signs of the runic fuþark alphabet (Hagland and Stenvik 2008). In addition to the example from Skei, runic inscriptions have also been identified on an Insular bronze ladle from Trå, Gardvin, in western Norway (Blindheim 1999, 51). Carefully made buckets carved from single blocks of yew with a separate disc of yew added to form the base comprise an even rarer group of artefacts among the Insular material. They are usually no taller than 100-190mm and are covered with thin bronze mounts richly decorated with Insular ornaments, bearing witness to an expensive manufacture. Based on their decoration, Bakka (1963, 32) has proposed a Northumbrian origin within the Hiberno-Saxon art province for the manufacture of these particular buckets. No more than seven of these are known from Norway, and three of them are found in Trøndelag; at Skei (Steinkjer), Halsan (Levanger), and Steinvik (Bjugn). They occur almost exclusively in women's graves primarily dated to the 9th century (Stenvik 2001, 29). While the buckets from Skei and Halsan were complete when found, the surviving remains of the Steinvik bucket comprise just two handle plates with small fragments of the original bronze mounts attached ( Figure 20). The function of such buckets in contemporary Norse societies has been little discussed, although it has been suggested that they may have been used to store or measure liquid, or perhaps materials such as flour and malt. Stenvik (2001, 29) has also proposed that they may have had an additional function as standardised measures of capacity.

Anglo-Saxon swords
Swords appear relatively frequently in the better equipped male burials in Viking-Age Norway. A small number of these swords, classed as Petersen type L, have been imported from Anglo-Saxon England. They are usually recognised by their triangular pommels with curving guards. In Norway, 18 swords belonging to this group are known, with marked concentrations in the counties of Vestfold and Nord-Trøndelag where 12 of the examples are located (based on information in Petersen 1919, Petersen 1940, Blindheim 1999, and searches in museum databases). The five type L swords from Trøndelag have all been found within the Trondheimsfjord to the north of Trondheim. Such swords generally occur in male burials dated to the 9th century, with occasional examples running into the following century (Petersen 1919).
Several researchers have in recent years highlighted how certain types of swords and other high-status weapons in Norse graves may have marked the buried person's place in society rather than purely reflecting an actual role as warriors or professional soldiers. Swords are likely to have been used as status symbols in all social environments, while the quality of both blade and guard were determined by the owners' social standing and economic capacity (Martens 1994;Andersen 2004). Owing to their small number and distinctive design, it is thought that swords of Petersen type L might have been utilised as particularly strong symbols of power and status among the upper social class in Norse societies (Andersen 2004, 99-100).
Traces of gold coating on the hilt of a type L sword from Sundnes, Inderøy, seems to confirm that Anglo-Saxon swords may have had other meanings besides their practical use.
Rare and imported weapons must have been treasured possessions and, therefore, sometimes repaired. A sword from Heggestrøa, Steinkjer, represents an example where the pommel and two silver mounts of a late Anglo-Saxon sword in Trewhiddle style have been reused and added to a new blade in the Viking period ( Figure 22). The main ornaments on silver bands attached to the grip consist of a series of contiguous lozenges. In the centre of each lozenge is a minute foliate motifusually quatrefoil, but occasional bifoliate. The ornaments are reserved against a niello background. A curved silver band above the pommel-guard is decorated in a similar fashion to the bands on the grip. Above this, the iron pommel is tripartite, divided by silver bands, while the centre of the pommel is embellished with a roundel, divided up into eight segments. The sword is a typical 9th-century weapon and can be compared in its embellishment and in the form of its pommel to two other Anglo-Saxon swords also found in Norway, those from Grønnesberg and Dolven, particular the latter which has lozenge and leaf pattern very similar to that on the Heggestrøa sword (Wilson 1965, 37). In addition to the above examples, only two other Trewhiddle style swords are known nationwide; the ones from Verdal, Larvik and Kaupang, Tjølling. This small assemblage of imported weapons thus appear as a rare and exclusive group among the large number of Viking-Age swords from Norway.

Other Insular finds
In addition to the groups described above, there are also five individual Insular finds of particular note: a complete reliquary shrine, a small silver button, a bronze bridle, an Anglo-Saxon disc-brooch, a jet bead and a part set of enamelled bronze mountings for a horse harness.
Although several of the bronze mounts found in Norway are thought to have originated as parts of reliquaries, only three complete shrines have ever been found in Norse burials in Norway (Heen-Pettersen 2013, 63). One of these shrines was found in a large grave-mound at Melhus, containing a boat burial where a man and a woman, usually believed to be husband and wife, had been laid to rest around AD 800. The reliquary is thought to have belonged to the woman. It is made of yew covered in bronze sheet, with a length of only 120mm and 80mm high. The lid is shaped like a roof with a high roof-tree and sloped gable-ends. On either side of the lid, there is a circular setting filled of tin plate, decorated with triskele ornaments (Figure 23 and Figure 24). The use and function of this shrine has been little discussed, although it has been suggested that it might have been the buried woman's personal 'keep-sake' box, to store small personal items such as jewellery, combs and the like (Heen-Pettersen 2013, 63). Reliquaries belonging to the same type as that from Melhus, characterised by having the shape of a small house with ridged roof and hipped ends, are known from Monymusk, Aberdeenshire, as well as Lough Erne and Shannon in Ireland, together with a fourth example currently preserved in the Copenhagen Museum. A small circular openwork bronze disc, identified by Bakka (1963) as a possible late Anglo-Saxon disc-brooch, was recovered from a woman's grave at Fulset, Stjørdal, in 1873 ( Figure  25). Traces of wear and refitting of the original pin show, however, that the brooch was of some age when put into the grave. The disc itself forms an equal-armed, Anglian cross with the arms splayed, so they touch one another and make the edge of the disc a closed circle. A close parallel to its form is provided by one of the Beeston Tor brooches (Bakka 1963, 23). The Fulset brooch was found together with two oval brooches of type P51, which generally date to the first half of the 10th century.  Figure 27). This burial was located close to the barrow where the previously discussed balance scale from Lø was discovered. The button is the only one of its kind found in Scandinavia, although several similar buttons have been found in Iceland, Ireland and the Isle of Man. Based on the design of the Lø button compared with known parallels, Grønnesby and Ellingsen (2012, 41) have suggested that this particular item may have been produced somewhere in the British Isles c. AD 920-960. Only a small number of artefacts of jet or jet-like materials (e.g. lignite) are known from burials in Norway (Shetelig 1944;Resi 2005). In Trøndelag only one such object, a bead, is recorded. The bead has, however, not undergone further analysis to establish whether it is made from jet or another material. The bead formed part of a woman's necklace, found in a double burial covered by a large grave-mound at Klingen, Overhalla. The nearest source of jet is Whitby, North Yorkshire, and black lignite shales are also commonly found as glacial erratics in the boulder clays of northern and eastern England (Shepherd 1985, 204). Artefacts of jet or similar material from Viking-Age burials are therefore usually considered as Insular imports (e.g. Hunter 2008; Resi 2005). Early types of bronze horse bridles are usually considered as Insular imports (e.g. Blindheim 1978;Petersen 1951, Jåtten 2006). One such bridle has been found in Trøndelag, at Nerrian, Skaun. Although found during ploughing, it probably came from a disturbed grave. Finally, Insular horse equipment is also represented by a set of enamelled bronze mountings from a man's cremation grave at Kolset, Steinkjer, dated to the 10th century. The set consists of three pieces of circular and square mounts, together with fragments of two twin-strap mountings. Although damaged by fire, several of the pieces bear remains of red enamel (Petersen 1940, 67-8). While a number of the reworked, bronze mounts found in Trøndelag are thought originally to have been part of horse harness, this is the only find where the mounts appear as part of a set. Traditionally, the beginning of the Viking Age has been defined on the basis of the first recorded attack on Lindisfarne in AD 793. Following this assault, the last decade of the 8th century saw an explosion of violence around the shores of Britain and Ireland, including raids throughout the Hebrides in 794, Skye and Iona, Inishmurray and Inishbofin in 795, mainland Scotland in 796, Ulster in 798 and Francia in 799 (Forte et al. 2005, 54). At the earliest, grave goods from these first recorded raids would have ended up in Norse burials in the years around AD 800. However, the nature of initial contact between Scandinavia and Scotland has been much debated, and it has been suggested that these first Viking raiders were drawn to the northern Insular areas as a consequence of direct overseas contact prior to these early recorded raids (e.g. Myhre 1993;Weber 1996). In such discussions, much attention has been paid to the identification of possible reindeer antler in pre-Viking type combs from the Orkney Isles. These combs have been put forward as evidence for a long and largely peaceful initial period of contact before the beginning of the recorded Viking raids, since reindeer in this period were not native to Britain, but widely distributed in Northern Scandinavia (Myhre 1993;Weber 1996.). However, applications of new techniques of raw-material analysis have demonstrated that the arrival of Scandinavian comb-making material in Atlantic Scotland occurred no earlier than the 9th century, and recent research has therefore dismissed this evidence for pre-Viking contact (Ashby 2009;von Holstein et al. 2014). The debate regarding the beginning of the Viking Age has nevertheless led to a more nuanced view of the first contact between Norway and the British Isles, although the first recorded attacks on the British Isles are still widely regarded as the beginning of the Viking Age. In Trøndelag, four female burials containing Insular artefacts from Grande (Ørlandet), Gjeite (Levanger), Skei (Steinkjer) and Melhus (Overhalla) may be dated to around AD 800, or perhaps slightly earlier ( Figure 28). Although their exact date is debated, the early types of brooches found in association with the Insular finds suggest that we are dealing with some of the graves that help to define archaeologically the beginning of the Viking Age (for further discussion, see e.g. Myhre 1993;Gaut 2001;Stenvik 2001). The bronze mount from Grande was found together with a pair of oval brooches of type R640, usually dated to the late 8th century (Jørgensen 2008, 32-33). The richly furnished cremation burial from Gjeite contained several Insular artefacts, together with two animal-shaped brooches generally assigned to the period immediately prior to the Viking Age (according to unpublished Vitenskapsmuseet catalogue information). One of the brooches showed signs of repair, however, and may have been of some age when put in the burial. A possible date of around AD 800 therefore seems reasonable for this find. An early date is also indicated by the find of a so-called Berdal type 1a brooch with gripping beast ornament from the Skei burial ( Figure  29). It has been suggested that such brooches began production in southern Scandinavia during the mid-8th century, based on workshop finds in Ribe, Jutland (Myhre 1993, 186). Stenvik (2001, 75) has suggested that the brooch from Skei may have come from this workshop and was put in the grave by around AD 775. There has, however, been considerable debate regarding the chronology of the culture layers associated with finds of moulds for Berdal brooches in Ribe. Based on excavations carried out during the early 1990s, Feveile and Jensen (2000, 17-18) have reassessed the Ribe dating evidence and argued that the date of the earliest occurrence of Berdal brooches must be adjusted to c. AD 790-800. The Insular gravefurniture from Melhus was found in association with two bowl-shaped brooches of the early type R643, generally attributed to the 8th century. They were accompanied by a particularly elaborate and large fibula brooch (type R649, Figure 30), which may have been produced in the Gotland area of Sweden. Such fibulas are also usually placed in the late Merovingian period, which in Norway is defined as AD c. 600-800. While some of these artefacts may well have been manufactured in the later 8th century, when they were deposited in the grave is, however, another question. Petersen (1907, 19-20) has therefore suggested that the brooches from Melhus were placed in the grave around AD 800. Early graves with Insular finds are traditionally interpreted as an indication of where the first Norse raiders set sail across the North Sea (e.g. Wamers 1998, 52). Some of the early foreign artefacts from Trøndelag can hardly be explained as a result of peaceful interaction. This is particularly illustrated by the reliquary shrine found at Melhus, together with a bronze mount that also appears have originated on an ecclesiastical object. It is unlikely that liturgical implements could be bought as common merchandise for export, and the grave goods from Melhus do seem to represent looted goods from one of the earliest raids on the British Isles. The grave goods from the other burials show a more nuanced picture of the earliest contact between Trøndelag and the British Isles. Insular material from Grande, Skei and Gjeite comprise both clerical and secular objects and seem to reflect a different form of contact than pure, random, plundered loot. The bronze mount from Grande has been identified as part of a horse harness rather than being part of a religious object (Wamers 1985, 93). Many of the Insular artefacts from the women's burials at Skei and Gjeite are serving equipment, for instance larger bronze vessels, ladles, buckets and drinking horns. Several of these artefact types are not generally associated with raiding monasteries and churches. In that connection, it is worth noting the remains of two drinking horn vessels from the Gjeite burial. According to literary sources, such items were considered appropriate gifts to one's allies (Youngs 1989, 74). Alliances formed between Norse chieftains and Irish kings are well recorded in written sources from a slightly later time (see e.g. Corrain 1972, 96;Crawford 1987, 47), and it is possible that the drinking horns from Gjeite are also examples of the same interaction. The number and composition of the early Insular grave finds seem to reflect that both raids and other forms of interaction with the local populations were operating, indicating wellestablished contact between Trøndelag and the British Isles at an early phase of the Viking Age.

The earliest contact
Perhaps not surprisingly, several of the earliest Insular finds from Trøndelag appear on farms where evidence shows that they were leading and powerful centres in the pre-Viking period. Gjeite, Levanger, stands out with a significant corpus of archaeological material for the period around AD 300-800, including gold rings, silver brooches and various bronze vessels imported from the Rhine area ( Figure 31). The richness of the farm may be explained by its strategic position. It lies by a natural harbour where communication lines from east, north and south meet, and a trading connection between the Levanger area and the inland areas of central Sweden may have been of particular importance. The wider archaeological evidence suggests that this trading contact may have already been established during the migration period (AD 400-600). Trade and exchange taking place in the Levanger area during the Viking period may well have been administered and controlled by a chieftain based at Gjeite (Herje 1989, 94). The cemetery at Skei borders the farm of Dalem, where some of Norway's richest grave-finds of early Iron Age date (AD 200-600) were discovered in the 1870s ( Figure  32). Stenvik (2001, 76) has proposed that Skei and Dalem once formed a single unit, representing an undocumented chieftain centre. This is supported by the presence within the cemetery of a distinctive tunanlegg (court site) of at least nine buildings radiating out around a central courtyard. Although their exact purpose is debated, they appear to have served both administrative and military functions (Stenvik 2001, 76). Finally, the region around Melhus, Overhalla, is an area with a marked concentration of richly furnished boat and stone cist burials from the early Iron Age through to the 10th century. The wider archaeological evidence indicates that this part of Namdalen at the beginning of the Viking Age occupied a prominent position among the northern districts of Norway (Sogness 1988). The importance of this region may be related to the location by the river Namsen and its tributaries. Namsen is a substantial waterway connecting several inland areas with the coast, making a wide range of resources from a number of different habitats easily accessible for trade and exchange. Control points along the banks of the river, administered by a few families, may have been important factors for obtaining economic recourses and power (Farbregd 1979, 67-68). Judging by the archaeological evidence and the farm locations, it seems likely that families from Skei, Gjeite and Melhus had the necessary political organisation, contacts and economic resources needed to organise overseas ventures at the very start of the Viking Age.

The 9th century: close relations between Trøndelag and the British Isles
During the first half of the 9th century, the quantity of Insular finds from Trøndelag increases significantly, and the majority of the whole assemblage can be dated to this century. The imported goods were, in all probability, acquired through various processes and patterns of interaction, and it can be somewhat difficult to separate different forms of commodity exchange in the archaeological material. From written sources, we know that in many cases specific circumstances decided whether one chose to exchange gifts, conduct trading or plunder, and the different forms of exchange cannot always be viewed as strictly separate spheres (Skre 2007). Innovation in maritime technology is often highlighted as an important cause of the increased Norse interaction with the British Isles during this century (e.g. Sørheim 2011;Roesdahl 1994). After the development of more advanced sea-going sailing ships, the British Isles were considered as a fairly close destination, which in fine weather could be reached from western Norway in a couple of days (Wamers 1998, 52). It is likely therefore that the Insular world was well known to the people living in coastal Norway, and that there may have been regular crossings to and from Trøndelag.
Insular finds assemblages of various sizes are found at several central places spread around the Trondheimsfjord, with marked concentrations at the areas of Skaun, Stjørdal, Levanger and Steinkjer. These places are located at natural harbours with good farmland, but, more importantly, strategically placed to control shipping and trade with products from further inland. Such central places, strictly controlled by local leaders and chieftains, functioned as local centres for trade, exchange and craft production and also served as collection points for local products for export from neighbouring settlements (Christophersen 1989, 121;Sogness 1988). In Trøndelag, the location of numerous inland iron production sites, major trapping systems for larger animals and locations of soapstone quarries suggests that iron, hunting products and soapstone are likely to have been particularly important local products for trade and exchange. These items would have been brought from inland to local central places by the fjord, where many of the goods changed hands and were transferred to different vessels before being sent out of the region (Stenvik 1994;Herje 1989.). Income from such activities is likely to have been an important factor in obtaining and maintaining sufficient resources needed to equip and organise long-distance voyages during this century. The distribution of Insular and other imported objects is often used to identify central places, since such finds can be viewed as an indicator of the abilities certain areas and communities had to equip and organise overseas expeditions during the Viking Age (Sørheim 2010, 275).
The area of Stjørdal, 50km north of Trondheim, is notable for the highest concentration of 9th-century Insular finds from Trøndelag. Within a distance of approximately seven miles, some 12 burials containing such objects have been found. The western part of Stjørdal has large areas of excellent farmland, while the eastern part is dominated by the valley of the Stjørdal River and its estuary. In addition to the fluvial deposits providing rich agricultural land, the valley would also have been an important route eastwards towards Sweden, making Stjørdal strategically placed for trade and exchange during the Viking Age. Many of the Insular finds from this area are clustered on neighbouring farms, and this is especially noticeable around the mouth of the River Stjørdal. The dense concentration of Insular material is striking, indicating that this part of Stjørdal may have played an important role in equipping and organising overseas journeys within Uttrøndelag. However, the wider archaeological material from Stjørdal implies that it may not have been a single, powerful chieftain centre here during the Viking Age, but rather a number of more-or-less equal aristocratic families within the same area (Winther 2011, 72-73). This is likely to mean that the power of each leading farm and chieftain was more limited here compared to some of the larger chieftain centres in other part of Trøndelag, such as Egge, Lade and Skei/Dalem, which show continued power over centuries. The cluster of Insular finds on neighbouring farms seems likely therefore to show that close collaboration between several families within the Stjørdal area was needed to be able to equip a fleet of necessary force. An extensive use of Insular symbols may in turn have been an effective way of expressing the area's authority and connection with overseas networks to the larger and more long-established chieftain centres in other parts of the fjord.
Although some Insular objects may have been traded or exchanged more widely into other parts of Norway, as noted, concentrations of this material are often considered to be an indicator of the points of origin of Vikings crossing over the North Sea (Sørheim 2011, 49). Based on the distribution of Insular finds nationwide, it is often highlighted that the areas of Rogaland and Sogn og Fjordane in the west of the country show the strongest links to the British Isles, with extensive migration from these areas (e.g. Bakka 1963;Wamers 1985). However, the significant increase in recognised Insular artefacts from Trøndelag now shows a similar number of burials containing this material in and around the Trondheimsfjord (45 locations) as in Sogn og Fjordane (49 locations, based on finds appendix in Jåtten 2006). The migration from Trøndelag to these new Norse areas is also illustrated dramatically by the discovery of a Viking-Age burial at Adwick-le-Street, South Yorkshire. Here, the grave of an adult woman was found, with grave goods comprising two oval brooches of later 9th-century date ( Figure 33, Figure 34). Finds of female Norse burials in England are rare and isotopic analysis was carried out to establish her place of origin. The result concluded that 'although a very specific area of north-east Scotland cannot be excluded as a possible childhood origin for this individual, the balance of probability from the combination of oxygen and strontium isotope data suggests Norwegian origin. The oxygen isotope composition of the drinking water only just falls in within the UK range and is a better fit with the Trondheim area of Norway' (Budd 2004, 63). Certainly, on the basis of the archaeological material, there is reason to propose there was a close relationship between societies in the Trondheimsfjord and Norse colonies in the British Isles, which developed during the course of the century. The link between Trøndelag and England can also be traced in the archaeological material itself. I cannot explore here the much-debated art history of early medieval Britain and Ireland, but will draw attention to the fact that, according to several studies attributing many types of Insular decoration to more-or-less specific art regions, only a relatively small number of metal objects from Norway are regarded as of English origin (e.g. Bakka 1963;Haseloff 1979;Wamers 1985). The distribution patterns showing that the Anglo-Saxon ornaments are, by and large, in the minority is seen as showing that the Norse interacted most frequently with the Celtic areas (Wamers 1985, 54). However, as already noted, a marked concentration of Anglo-Saxon artefacts has been found at Stjørdal, where half of the 12 burials with Insular material contained Anglo-Saxon metalwork, including two swords, three bronze mounts and a disc-brooch ( Figure 21, Figure 25, Figure 35, Figure 36)

The 10th century; towards centralised control over the western contact?
From around AD 900, the datable Insular material from Trøndelag shows several changes in the nature of contact. Within large parts of the Trondheimsfjord, there is a sudden disappearance of Insular finds in areas that appeared to have frequent contact in the previous century, suggesting that the westward ventures had come to an end for many of the communities within the fjord. This is particularly notable in Stjørdal, where the preceding strong connections more-or-less seem to vanish around the beginning of the 10th century. The decrease of datable finds not only from Stjørdal, but also in other parts of Uttrøndelag, is significant. Part of the decline is almost certainly related to changed conditions within the Norse settlements in the British Isles, where the activity and the evolving communities in the later 9th century became more politically controlled and where 'there was no place for old style war bands under individual chieftains raiding monasteries and looking for land to settle' (Crawford 1987, 63). However, while the number of Insular finds declines significantly in Uttrøndelag, finds of such artefacts seem to continue to some extent in Inntrøndelag and Namdalen, indicating additional regional and local factors may have been at work.
Wealth acquired by warfare and trade was an effective means of winning support and building up power and prestige within the social system of Norse communities (Helle 1991, 26). The economic and ideological resources that could be drawn from the Insular networks must therefore have been significant, providing a considerable source of income and status for those involved. Access and influence over the North Sea activity would also have given the local elites strategic access to the aristocratic networks of northern Europe (Bjørgo 1995, 27). The networks with the Insular world would therefore have represented both an essential economic and ideological prerequisite for maintaining and strengthening the power base for certain groups. Control over trading and exchange became of increased importance towards the end of the Viking Age as both local and long-distance trading increased in volume (Skre 2007). Heimskringla refers to an increased consolidation of power in Trøndelag during the last part of the Viking Age, resulting in conflicts between powerful Norse leaders who were seeking to establish and control trade within the Trondheimsfjord. This is likely to have led to increased control of North-Sea activity, which in turn is liable to have been another factor diminishing the opportunities for individual chieftains from smaller communities around the fjord to organise westwards voyages. It may therefore be no coincidence that the overall finds distribution within the Trondheimsfjord area indicates that westward contact, which still existed in the 10th century, was confined to Inntrøndelag, especially to the area around Steinkjer. This may be related to the emergence of a central trading place at Steinkjer towards the end of the Viking Age as indicated by written sources and archaeological finds (Grønnesby and Ellingsen 2012, 43-44). According to the Saga of Olav Haraldsson (St Olaf), the Jarl of Lade -Eirik -founded a trading place (Kaupang) at Steinkjer in an attempt to make Steinkjer the leading trading centre in Trøndelag at the expense of the royal centre at Nidaros (Trondheim). This is likely to have been a policy both to limit increasing royal power and Christian influence in the region and to reinforce the Jarls' power base in Inntrøndelag, where the old faith and traditions were still strong (Skjevik 1997, 146).

The Steinkjer
Kaupang is supposed to have been of greater importance than Nidaros for a short period before King Olav Haraldsson again restored Nidaros as the royal residence and trading centre, towards the end of the Viking Age. The distribution of imported finds in that area indicates that the Kaupang in Steinkjer was in operation during parts of the 10th century, although its exact location remains uncertain (Grønnesby and Ellingsen 2012, 31, 44). Archaeological evidence from burials at Lø in Steinkjer, however, suggests that this farm may have had a central role in the control and administration of trade and exchange during this period, possibly on behalf of a leading chieftain in Inntrøndelag (Herstad 2012, 104-5). The Insular material from Lø includes the balance scale previously discussed, and it is worth noting that this find, together with a balance scale from the neighbouring farm Egge (of possible eastern origin), so far represent the only weighing equipment found in 10th-century graves from within the Trondheimsfjord.

Conclusion
This review of Insular grave finds from Trøndelag has shown that there is significantly more material from this region than has previously been recognised. Imported items include both religious and secular objects of great rarity, value and beauty. In terms of the quantity and quality of Insular artefacts from Norse graves, Trøndelag can be shown to match several of the 'traditional' areas of Viking contact in Western Norway. The artefacts conjure up a picture of a life where overseas voyages, probably both raiding and trading, played an important part for many communities within the region. Within their societies, the Trøndelag Vikings' affiliation with Ireland and Britain was visually displayed through the use of exclusive dress ornaments, for example the outstanding ring brooches from Snåsa and Nes, and by foreign exotica brought home for festive use on the high table, as illustrated by the drinking and serving equipment from Steinvik, Gjeite and Skei. The diverse corpus of Insular importstogether with their Scandinavian imitations -can be seen as evidence of the creation of a marked 'Insular milieu', which had a definitive and lasting influence on the daily life and the mentality of the Scandinavian population (Wamers 2011, 97).
The earliest evidence of contact between Trøndelag and the British Isles is represented by a group of especially richly furnished women's graves dating to around AD 800, or perhaps slightly earlier. The archaeological material from these burials and their nearby surroundings indicates that the communities around the Trondheimsfjord had the necessary political organisation, contacts and economic resources needed to organise overseas ventures during the initial phase of the Viking Age. Previous work on early Insular material from Norway has, however, been broad and synthetic (e.g. Wamers 1985). A detailed review of the material from the graves in Trøndelag is therefore a key contribution to the wider debate about the start of the Viking Age.
The main evidence for contact comes from the 9th century, when a number of significant patterns can be discerned. Some local concentrations of Insular goods show the continuing importance of some pre-Viking centres, while the finds distribution in other areas suggests cooperation between several neighbouring families in order to equip and provision overseas expeditions. Later contact and North Sea trading appear to be affected by central control as a result of increased centralisation of power in Trøndelag during the 10th century. This is particular notable in the Trondheimsfjord, where the Insular contact in the latter part of the Viking Age is confined to the area around Steinkjer in Inntrøndelag. This might be related to the emergence of a trading place (Kaupang) at Steinkjer, which was established in an attempt to control trade within the Trondheimsfjord. The Kaupang in Steinkjer is, however, thought to have been short-lived and towards the end of the Viking-Age Nidaros was restored as the kings' residence and trading town. According to the saga of Olav Haraldson (St Olaf), Eirik Jarl, who founded the Steinkjer Kaupang, left Trøndelag in AD 1015 for England, becoming earl of Northumbria, where he died (Skjevik 1997, 156-7). Only a few years later, the fall of Olav Haraldson in battle at Stiklestad in AD 1030 conventionally marks the end of the Viking Age in Norway. By this time, Insular and indeed other finds had ceased to be buried in graves in Trøndelag, owing to the increasing influence of the Christian faith replacing the pagan traditions of accompanied burial. Nidaros was now the leading centre of Trøndelag and would later become the medieval metropolitan capital of Norway, with a cathedral containing the relics of the martyred king, St Olaf.