Våben i vand – Om deponeringer i vikingetiden


  • Julie Lund




våben, deponering, vikingetid


Weapons in water

The purpose of the article is to throw light on Viking Age weapon finds from wetland areas. In the presentation, the author claims that these weapons should be interpreted as traces of ritual acts of deposition rather than as lost items or traces of actual battles. The positions of the weapon deposits in the landscape are discussed. Further, the interpretation of a number of written sources mentioning weapons in connection with wetlands is discussed. An examination of Viking Age weapons found in wetlands in Zealand and Scania shows that they are concentrated in the small rivers of Værebro Å, Tude Å, Lavringe Å, Dybecks Å, Sege Å, and Kävlinge Å; in the Borremosen bog on Møn; in the lakes of Tissø, Näsbyholm Sjö, Søborg Sø, and Højby Sø, and in the wetlands surrounding Kristianstad (Fig. 1).

The weapons found in the river Værebro Å (Fig. 2) are discussed. Traditionally, these weapons were interpreted as the remains of a battle that took place in 1133. However, the majority of the weapons are from the 10th century. In addition, both jewellery and tools have been found in the river, and these finds cannot be interpreted as remains of actual battles. The author therefore suggests that the finds from Værebro Å should be interpreted as ritual deposits. In addition, numerous Viking Age weapons have been found in wetlands in Sweden, in the British Isles, and on the Continent. These finds are usually interpreted as the remains of ritual deposits, and the same interpretation is therefore suggested as applying to the Danish and Scanian material.

Seen as a whole, the material from Zealand and Scania tends to accumulate near the mouth of small rivers and streams (Fig. 3), and in natural harbours. There is also a tendency towards weapon deposits accumulating near bridges and fords (Fig. 4). It is shown that weapons from the Middle Ages tend to be concentrated in wetlands near castles, and these finds often include a few Viking Age weapons from the time before the building of the castle. Weapon finds from Dybecks Å and Näsbyholm Sjö are examined. The lake of Näsbyholm Sjö contained axes from the 9th to the 12th century (Fig. 5). The river Dybecks Å runs out of the lake, and in it a sword, decorated in the English style, was found near the Herremandsbro bridge, next to which a rune stone called the “Östervemmenhögssten” was raised in memory of “a very noble man”. This type of stone was often raised in commemoration of men that died on Viking expeditions. The sword and the rune stone both date from the late 10th century.

The author analyses a number of written sources from the late Viking Age and the early Middle Ages. The poem Vóluspa 21 from the older Edda describes “a stream called Slid, flowing with swords and saxes.” This stream is also mentioned in Grimnesmal 28 from the same Edda, which describes several streams that “run through Midgard, but from here they fall to Hel.” Some of these streams have names that reflect fear, battle, and death in combat. In addition, a new translation of Adam of Bremen’s scolie 138 is suggested: instead of translating the word fons into a “brønd” (well), it should be translated into a “kilde” (spring). If so, Adam of Bremen describes the pagan custom of making sacrifices to springs in Uppsala, which is in keeping with the incidence of weapons from the 9th to the 13th century found in the river Fyris Å, which flows through Uppsala.

In some of the rivers there is a long continuity of weapon deposits. This applies to the finds from Tissø, Værebro Å, Søborg Sø, Højby Sø, Näsbyholm Sjö, and from the wetlands surrounding Kristianstad (Figs. 6 & 7).

The author suggests that the weapon deposits are connected to overseas expeditions. This interpretation is based on the fact that large quantities of the weapons found are in harbours or river-mouths. In addition, many Viking weapons have been found in areas of the Continent and the British Isles where the Scandinavians stayed for a prolonged period. The importance of the expeditions during the Viking Age is seen in the fact that around one third of the Swedish rune-stones describe men that died on expeditions, and these rune-stones were often placed near large waterways and major roads.

The author also suggests that the weapon deposits near bridges and fords could be explained by the importance that these special landscape features had. The fact that bridges had considerable cultural importance is stressed by the large number of rune stones that were erected next to bridges towards the end of the Viking Age. As weapon deposits near bridges occur during the whole of the Viking Age, this may mean that the building of a bridge should not be understood as a Christian act, but as an act of ritual or religious importance that continued from the pagan era into Christian times. In Gylfaginning 48 from Snorre’s Edda, Midgard and Hel are separated by a river, and a bridge leads to the entrance to Hel. The rivers described in Grimnesmal also indicate that some rivers lead to the realm of the dead. There thus seems to be a connection between death and certain waterways. Perhaps this explains why bridge-building became an important Christian act during the late Viking Age and early Middle Ages. The author finally suggests that weapon deposits should be seen as a manifestation of social identity, as deposits in wetlands during the Viking Age follow the same groupings that we know from grave goods such as weapons, jewellery, and tools.

Viking Age weapons found in wetlands thus seem to be the remnants of ritual acts, rather than the remains of battles. The finds show that wetlands and waterways were significant features in the Viking Age countryside. Not only were they important means of communication and travel, they were also of fundamental importance to man’s self-concept and philosophy of life.

Julie Lund
Institut for Arkæologi og Etrologi
Københavns Universitet

Translated by Annette Lerche Trolle





Lund, J. (2004). Våben i vand – Om deponeringer i vikingetiden. Kuml, 53(53), 197–220. https://doi.org/10.7146/kuml.v53i53.97499