The following article presents a preliminary publication of a shield that was discovered in 2010 in a ship burial from Ness, Norway. The presented text, which substitutes the currently missing detailed academic publication, uses photographic material taken in the permanent exhibition of the The Arctic University Museum (Norges arktiske universitetsmuseum) in Tromsø, a description available on the server Unimus.no (Unimus 2023) and existing literature. The findings are compared with the closest analogies that were already published in our summary article Lesser known aspects of the Viking shield.
The position of Ness on the map of Europe.
Introduction of the grave and inventory
In 2010, under the leadership of J. E. Arntzen, rescue archaeological research of a damaged burial mound in Ness, Hamarøy commune (Nordland region) began, which continued in 2011 after extraordinary discoveries. The research discovered a looted inhumation ship burial and a remarkable structure in its immediate vicinity (Arntzen 2011; 2015). The 10-12 meter long ship, oriented approximately in the north-south direction, from which more than 100 rivets have been preserved, rested in a pre-dug ditch and the sides of the ship were lined with stones (Eldjarn 2021). The body of a man aged 50-60 was placed in the center of the ship. The deceased was apparently placed on a coarse woolen textile. An iron anchor, approximately 100 cm long and weighing 10 kg, was placed on his left side, while a large cauldron with a diameter of 40 cm was found to the right of the deceased, placed upside down. The deceased’s clothing included both a tablet woven band of silk and silver wire, exposed on the back of the skull (Storli 2021), as well as braided silver wire application, a 20 cm long silver ringed brooch at the thigh, and a pair of Carolingian giled buckles with matching strap ends. Two shields were placed at the feet of the deceased. Other equipment included a weight, an arrow, a horse bit, beads and probably also a comb. Numerous finds of deposited animal bones show that many animals were sacrificed during the burial, including three horses. A mound with a diameter of 23 meters was built above the grave, which was surrounded in close proximity by a palisade of at least 34 columns made of sturdy logs. The objects are currently stored in the The Arctic University Museum in Tromsø under the inventory number Ts12156.
The mound at Ness and its graphic representation. Source: Arntzen 2015.
The dating of the grave varies in the published literature. Arntzen in one place suggests that the grave is older than 1100 years, thus placing the dating in the 9th century (Arntzen 2015: 29), and in another place he dates the shield bosses of type R562 between 850-950 (Arntzen 2014). Inger Storli dates the grave to the year 900 (Storli 2021) and Marianne Vedeler leans towards the 10th century (Vedeler 2014: 42). We can agree with the general dating to the second half of the 9th or the first half of the 10th century on the basis of shield bosses (Hjardar – Vike 2016: 185). The long spatula-shaped shield handle that is present in the grave is rather a feature of the first half of the 10th century, as we have shown in the article The shield handle from Myklebost. An illustrative example in the Norwegian area is the find from Gokstad, which is precisely dated to the year 905 (Hjardar – Vike 2016: 185). Silver wire balls and plain weavings appear in the Anglo-Saxon environment already in the last third of the 9th century (Graham-Campbell 2013: 122-3), but in the context of Scandinavian graves we see them from the beginning of the 10th century (Steinforth 2015: 51; Thorvildsen 1957: 79-81). An interesting and hitherto overlooked chronological anchor of the Ness grave is a pair of Carolingian giled garter fittings, which are among the unusual objects in Scandinavia. It is a continental production typical of the 9th century (e.g. Kouřil 2014: 388-390; Šolle 1966: Tab. XXV), while the closest analogy in Scandinavia is the find from Ladby, which can be dated to around 925 (Thorvildsen 1957: Taf. 1). Putting these data together, the 1st half of the 10th century seems a more likely choice for dating the Ness grave, which in its magnificence ranks close to the group formed by the ship graves from Gokstad, Ladby and Myklebost. It should be noted that the grave from Ladby is the only known parallel to a ship burial surrounded by a palisade (Sørensen 2001: 36-7; Thrane 1987).
Video about the find context. Source: Ragnar L. Børsheim, Vimeo.
Shield description and analogies
From the available information, it is clearly evident that the deposited shields belonged to the most luxurious class of this type of militaria. Due to the absence of a detailed grave plan, it is not possible to determine whether both shields were richly decorated or if the expensive decoration was only on one shield and the other was simpler. In the following description, we approach the problem by merging all preserved elements into one shield.
The base of the shield was a wooden board, of which only fragments have been preserved in connection with the metal edge and the boss. The number of boards is usually between 6 and 8. The material of the shield from Ness has not been analyzed, but the wooden corpus is usually made of conifers in Scandinavia (Arwidsson 1986: 39; Dobat 2013: 164; Christensen 1981: 100; Nicolaysen 1882: 62; Urtans 1961: 222; Warming et al. 2020: 172, 174), less often from deciduous trees (Pentz 2009: 170-1; Warming 2016). The diameter of the board cannot be determined other way than on the basis of the length of the handle. The thickness of the board in the center can be determined approximately at 0.8 cm based on the rivets of the boss. It can be assumed that the edges of the board were tapered as in the case of analogies; from the available photos, it seems that the thickness of the edges does not deviate from the average (around 0.4-0.6 cm).
According to the description available in the Unimus catalog, the board is covered with hide, which can be expected on both sides (Warming et al. 2020). It is usually a thin and transparent parchment-like material. Due to the board size, two skins of smaller animals (for example sheep) are sometimes used to cover each side of the corpus, which are sewn together and to the corpus at the same time (Warming 2023; Warming et al. 2020: 169); however, we have no idea whether the shield from Ness possessed this feature. According to the description, the hide bears traces of painting (Arntzen 2015: 30), which may have consisted of “red-orange colour” and “white and black dots”. As far as we know, no pigment has been analyzed. Red, white and black colours are among the popular colours for shields, which can be found separately or in combination in finds from Ballateare (Bersu – Wilson 1966: 58-61), Gokstad (Nicolaysen 1882: 62-3), Gamla Uppsala and Gnezdovo (see Lesser known aspects of the Viking shield), Trelleborg (Dobat 2013: 164), Tranås (Svanberg 2003: 297) and an atypical, shield-like object from Grimstrup (Stoumann 2009: 33-40) and are mentioned in written sources (Falk 1914: 128-132) .
The shield bosses from Ness. Source: catalog Unimus.
The center of the visible side of the shield was fitted with an undecorated boss of type R562 protecing the hand. Bosses of this type are very common and in the Viking world we know probably hundreds or even thousands of them (Hjardar – Vike 2016: 185; Petersen 1919: 47). Its size is slightly oval, roughly 15.5×15.2 cm with a flange width of 1.2-1.5 cm. The height of the boss reaches approx. 7.5 cm. It is noteworthy that, while usually the flanges are riveted to the boards with an even number of rivets or in an even number of positions (4 or 6), in our case there are apparently 11 rivets, two of which also fixed the handle at the same time. The number of rivets is interesting because, in the case of an even number, the handle (which is attached by two opposing rivets) bisects the hole symmetrically, while an odd number of rivets indicates that the handle may be placed asymmetrically (with an odd number of rivets and an asymmetrically placed handle, it is likely that the shield can be grasped in just one way). In the case of the Ness boss, it is possible that the handle bisected the slightly oval opening into two roughly similar halves. The back side of the rivets is extended into flat surfaces with edges (pads?), the exact nature of which we are unable to estimate from the photographs.
An outstanding part of the find is a decorated handle, preserved in at least six fragments (33, 226, 227, 232, 337, 338) with a minimum total length of 55 cm. It can be assumed that the original length was close to 70-80 cm. The core of the handle is formed by a concave one-piece iron plate, which is finished with flat, spatula-shaped ends about 6-7 cm long. The entire outer part of the handle was originally covered with 2.5 cm wide stamped strips of gold-plated copper alloy – the manufacturer originally made one long strip, which he decorated with punches and from which he cut shorter strips as needed. The strips are decorated with relief circles, triads of dots and perimeter lines. The strips are bent into the interior of the metal core by a few millimeters, where they were held by a wooden filler. Spatula ends were covered with wider strips decorated with a different ornament – a man’s face and a braid. At the narrowest points of the handle, between the last standard strips and the spatula-shaped ends, there are atypically narrow strips decorated with braid. The widest point of the coated handle (center) is approximately 2.5 cm wide and 1.5 cm high at the base, while the narrowest point in front of the spatula ends is approximately 1.5 cm wide and lower than the center. We can expect the thickness of the sheet to be roughly 0.15 cm, as in the case of analogies. The handle was attached to the board with iron rivets, which, with the exception of the spatula ends, pass through the coating. The number of rivets located under the cover of the spatula-shaped ends is unknown. It is not known whether the shield had a hanging mechanism.
Remains of the handle from Ness and its ornament. Source: Arntzen 2011: Fig. 7; catalog Unimus.
Details of handle deformations. Source: Jan Kozák; Christel Meyer.
The author of this article collected no less than 53 metal remains of Scandinavian shield handles (Lesser known aspects of the Viking shield), which spontaneously divide into two groups:
1. handles fixed with three-leaf terminals made of copper alloy.
The core can be a wooden handle or a metal tube with a semicircular cross section. In the case of the metal variant, decorated strips, inlayed or plated wires are usually applied to the tube. The ends of the handle create sharp spikes, which are covered with three-leaf terminals. The distribution of this variant lies in the Baltic region and in Old Rus. Handles with three-leaf handles are generally very short, around 30-40 cm.
2. handles with extended, flattened ends.
The core is always a metal tube with a semicircular cross-section, which tapers towards the ends, but the ends themselves are widened into spatulas, which are riveted with two rivets and usually subsequently covered with a copper alloy sheet to hide the rivets. The entire surface of these handles is usually covered with copper alloy strips. We are see the distribution of this variant evenly in Norway, Sweden and Denmark. Handles with extended ends usually copy the entire diameter of the shield.
Distribution of short metal handles with terminals (blue) and long handles with extended ends (orange).
The closest parallels to the handle from Ness include finds from graves Bj 736 and 504/536 from Birka (Arbman 1940: Taf. 19.9; Arbman 1943: 146-147, 259-261, Abb. 92-93), Rends (Brøndsted 1936: 122-3; Eisenschmidt 2004: 399; Pedersen 2014a: 100-1; 2014b: 98-99, Pl. 44), Myklebost (Schetelig 1903: 6; 1905: 14-6; 1912: 168) and Gokstad (Nicolaysen 1882: 49, Tab. X.20), which are characterized by a metal base covered with a sheet of copper alloy. For the sake of completeness, it is also possible to mention the finds from grave Bj 561 from Birka and the settlement find from the same locality, which represent undecorated spatulate ends of handles (Arbman 1940: Taf. 19.8; 1943: 180-1; Sörling 2018: 31). The circular motif on the handle strips from Ness is unique, but in terms of spatula-shaped ends, the closest ornament can be found in graves Bj 736, Rends, Myklebost and Gokstad. Shields of this style are similar enough that they may come from the identical series. It is also worth noting the fact that grave Bj 736 in Birka also contained two shields, which in their overall appearance show striking similarities to the shield from Ness. The dating of this grave also goes to the first half of the 10th century.
Edge fittings of the shield from Ness. Source: Jan Kozák; Christel Meyer.
The last significant component is the edge of the shield itself, which has been preserved in at least nine fragments (341, 349, 393, 396, 531, 543, 544, 550, 554) and which was partially or completely surrounded by a continuous layer of thin and in some places probably decorated sheet metal made of a copper alloy, fastened at regular intervals with iron clamps. This sheet had a U-shaped cross-section and was approximately 1 cm high when folded. The size of the clamps is close to 2 × 2 cm and their cross-section is similar to the shape of a keyhole – the upper part has been widened to accommodate the copper alloy rim, while the lower part is pressed to the level of the hide-covered board. A pair of iron rivets are located on this lowered part. The clamps are spaced close to 1 cm apart, allowing the copper alloy edge to shine through. According to the description in the Unimus catalog, the clamps also bear traces of decoration, which, however, are not visible in the photos.
The edges of Scandinavian shields of the Viking Age were typically edged with a leather band that was glued on (best example see Bj 850 in Birka; Warming et al. 2020) and apparently also partially sewn (Warming 2023). The main purpose of the edge is to provide structural strength to the shield, prevent deep penetration of the shield, and protect the exposed organic layers covering the shield. Metal rims are very unusual and potentially include finds from Laxare (Thunmark-Nylén 2006: 319), Tuna Alsike 4 (Arne 1934: 30, Taf. X.6) and Ivla Ödegård (Svanberg 2003: 192), with the Ness find being the only example from the Viking Age that has a cross-section of the letter U. Analogical solutions can be found in Celtic (Stead 1987), Roman (Sim – Kaminski 2012: 148) and Germanic shields (Arwidsson 1977: Abb. 49, Taf. 6-7; Bruce – Mitford 1978; Ilkjær 2001: 335-7). We can find occasional mentions of iron rims in Sagas of Icelanders, for example Grettis saga (ch. 40) and Kormáks saga (ch. 12) (Falk 1914: 139-140). As for the metal clamps, this is the more numerous component, having been found in at least 58 situations in the Scandinavian world. The purpose of clamps is either to hold the edge around the entire circumference, to repair the edge, or to secure the edge where the ends overlap. Clamps, which are found on the shields in the number of 1-60 pieces, are overwhelmingly made of iron, less often of copper alloy. Some clamps are tinned (Arwidsson 1986: 43) or even gilded (Owen – Dalland 1999: 132, Fig. 91). Additional decoration with side lines, rows of stamped dots or pits surrounding the rivets is not uncommon. The number of rivets in the clamps is almost exclusively equal to two, exceptionally three (Ramskou 1976: 38-9, Fig. 92). It is not clear whether the shield from Ness was rimmed only on part of the edge (analogy: grave Bj 628 from Birka) or whether it was rimmed around its entire circumference (analogy: grave L-206 from Gnezdovo, grave Bj 736 from Birka).
Distribution of edge shield clamps made of copper alloy (orange), iron (blue) or unknown material (black).
Conclusion and acknowledgments
Our preliminary shield publication introduced the Ness ship burial, offering a revision of the dating and a basic view of the deposited shields, which belong to the most prestigious class of Viking Age shields; shields of these qualities became gifts, were praised in poems, and finally accompanied early medieval magnates to the afterlife. In the post, we presented analogies of the individual components, including the handle and clamps. The closest analogies of the handle come from Norway (Gokstad, Myklebost), Sweden (graves 504/536 and Bj 736 in Birka) and Denmark (Rends), and their similarity is so striking that it would be theoretically possible to talk about one workshop. Clamp distribution suggests that these components are not common in Norway, unlike Denmark and Sweden. The sheet metal edge with a U-shaped cross-section is a unique feature that has no parallels in the Viking Age. Based on the parallels of these metal components, one can rather lean towards the opinion that the shield was imported from the more southern part of Scandinavia. The dating of the better equipped graves with similarly decorated specimens leads to the first half of the 10th century.
It must be emphasized that the study of early medieval shields is a changing field and, considering the low number of specimens, we have significant gaps in the recognition and comprehensive understanding of individual elements, which we fill with new finds and hypotheses that may later prove to be false. A detailed description of each new discovery is therefore an extremely valuable contribution to the discussion that shapes this dynamic field.
The article would not have been possible without the help of colleagues Jan Kozák (Charles University, Prague) and Christel Meyer (The Arctic University of Norway, Tromsø), who selflessly took photos of unpublished edge fittings for us. A big thank you to them for that. We also thank Are Pedersen for consulting the obtained material.
Schematic drawing of the assumed look of the shield from Ness.
Author: Michal Havelka, baba_jaga_atelier. Better resolution here.
We hope you liked reading this article. If you have any question or remark, please contact us or leave a comment below. If you want to learn more and support our work, please, fund our project on Patreon, Buymeacoffee or Paypal.
Arbman, Holger (1940–1943). Birka I. Die Gräber. Text (1943), Tafeln (1940), Stockholm.
Arntzen, J. E. (2011). En «palisadehaug» fra yngre jernalder i Nord-Norge? In: Nicolay. Arkeologisk tidsskrift 114, 33-42.
Arntzen, J. E. (2014). Vikinggraven på Ness, Hamarøy. In: digitaltmuseum.no [online]. [2023-02-23]. Available here: https://digitaltmuseum.no/011085442248/vikinggraven-pa-ness-hamaroy.
Arntzen, J. E. (2015). Historia om en gravhaug fra vikingtid på Ness i Hamarøy. In: Ottar 305 – 2015/2, 28-33.
Arwidsson, Greta (1977). Valsgärde 7, Uppsala.
Arwidsson, Greta (1986). Schilde. In: Arwidsson, Greta (ed.). Birka II:2: Systematische Analysen der Gräberfunde, Stockholm, 38–44.
Bersu, Gerhard – Wilson, D. M. (1966). Three Viking Graves in the Isle of Man. The Society for Medieval Archaeology Monograph Servies: No 1, London.
Bruce-Mitford, Rupert (1978). The Sutton Hoo ship-burial. Vol. 2: Arms, armour and regalia, London.
Brøndsted, Johannes (1936). Danish inhumation graves of the Viking Age. In: Acta Archaeologica VII: 81–228.
Dobat, A. S. (2013). Kongens borge: Rapport over undersøgelserne 2007-2010. Jysk Arkæologisk Selskabs Skrifter 76, Århus.
Eisenschmidt, Silke (2004). Grabfunde des 8. bis 11. Jahrhunderts zwischen Kongeå und Eider : zur Bestattungssitte der Wikingerzeit im südlichen Altdänemark, Neumünster.
Eldjarn, Gunnar (2021). Båtene fra Ness i Hamarøy og Hillesøy i Tromsø. In: Ottar 334 – 2021/1, 16–21.
Falk, Hjalmar (1914). Altnordische Waffenkunde, Kristiania.
Graham-Campbell, James (2013). The Cuerdale Hoard and related Viking-Age silver and gold from Britain and Ireland in the British Museum, London.
Stoumann, Ingrid (2009). Ryttergraven fra Grimstrup : og andre vikingetidsgrave ved Esbjerg, Ribe.
Hjardar, Kim – Vike, Vegard (2016). Vikings at War, Oxford – Philadelphia.
Christensen, A. E. (1981). Skjoldene på skipet = The shields of the Gokstad ship. In: Wexelsen, Einar (ed.). Gokstadfunnet : et 100-års minne = The Gokstad excavations : centenary of a Norwegian Viking find, Sandefjord, 16-7, 91-2.
Ilkjær, Jørgen (2001). Illerup Ådal 10. Die Schilde (Katalog, Tafeln und Fundlisten), Højbjerg.
Kouřil, Pavel (2014). Great Moravia and the beginnings of Christianity, Brno.
Nicolaysen, Nicolay (1882). Langskibet fra Gokstad ved Sandefjord = The Viking-ship discovered at Gokstad in Norway, Kristiania.
Owen, Olwyn – Dalland, Magnar (1999). Scar: a Viking boat burial on Sanday, Orkney, Tuckwell.
Pedersen, Anne (2014a). Dead Warriors in Living Memory. A study of weapon and equestrian burialsin Viking-age Denmark, AD 800-1000, Publications from the National Museum. Studies in Archaeology & History Vol. 20:1 1. (Text), Copenhagen.
Pedersen, Anne (2014b). Dead Warriors in Living Memory. A study of weapon and equestrian burialsin Viking-age Denmark, AD 800-1000, Publications from the National Museum. Studies in Archaeology & History Vol. 20:1 2. (Catalogue), Copenhagen.
Pentz, Peter (2009). Bådgraven ved Flintinge Å – og en kammergrav ved Rabjerg. In: Aarbøger for nordisk Oldkyndighed og Historie 2005, 161-208.
Petersen, Jan (1919). De Norske Vikingesverd: En Typologisk-Kronologisk Studie Over Vikingetidens Vaaben, Kristiania.
Ramskou, Thorkild (1976). Lindholm Høje : Gravpladsen, Copenhagen.
Schetelig, Haakon (1903). Fortegnelse de til Bergens Museum i 1902 indkomne sager ældre end reformationen. In: Bergens Museums Aarbog 1903, No. 3, 3-39.
Schetelig, Haakon (1905). Gravene ved Myklebostad paa Nordfjordeid. In: Bergens Museums Aarbog 1905, No. 7, 3-53.
Schetelig, Haakon (1912). Vestlandske graver fra jernalderen, Bergen.
Sim, David – Kaminski, Jaime (2012). Roman Imperial Armour: The Production of Early Imperial Military Armour, Oxford.
Sörling, Erik (2018). Fynden från ”Svarta jorden” på Björkö : från Hjalmar Stolpes undersökningar, Katalog, Uppsala.
Stead, I. M. (1987). The Chertsey shield. In: Surrey Archaeological Collections 78, 181-3.
Steinforth, D. H. (2015). Die Wikingergräber auf der Isle of Man, Oxford.
Storli, Inger (2021). Godbiter fra arkeologisk magasin – tre spesielle tekstiler fra jernalderen. In: Ottar 337 – 2021/5, 35-47.
Svanberg, Fredrik (2003). Death Rituals in South-East Scandinavia AD 800-1000 : Decolonizing The Viking Age Vol. 2, Stockholm.
Sørensen, A. C. (2001). Ladby: A Danish Ship-grave from the Viking Age, Roskilde.
Šolle, Miloš (1966). Stará Kouřim a projevy velkomoravské hmotné kultury v Čechách, Praha.
Thorvildsen, Knud (1957). Ladby-skibet, København.
Thrane, Henrik (1987). The Ladby ship revisited. In: Antiquity 61 (231), 41–49.
Unimus 2023 = Ts12156. In: Unimus [online]. [2023-02-23]. Available here: https://www.unimus.no/portal/#/search/things?freetext=Ts12156&requirePhoto=false&limit=100.
Urtans 1961 = Уртан, В. А. (1961). Древние щиты на территории Латвийской ССР // Советская Археология, 1, 216-224.
Vedeler, Marianne (2014). Silk forthe Vikings, Oxford – Philadelphia.
Warming, Rolf (2016). Shields and Martial Practices of the Viking Age: Shield Finds from Viking Age Denmark and the Functional Aspects of Round Shield Constructions. Copenhagen: University of Copenhagen.
Warming, R. F. (2023). The shields from the Viking Age ship burial at Gokstad: A re-examination of their construction and function. In: Arms & Armour 20/1.
Warming, R. F. et al. (2020). Shields and hide. On the use of hide in Germanic shields of the Iron Age and Viking Age. In: Germania 97, 154-225.