Kærgård ved Daugbjerg – Bebyggelse med værkstedsområde fra yngre jernalder


  • Kamilla Fiedler Terkildsen
  • Marianne Høyem Andreasen




Kærgård ved Daugbjerg, bebyggelse, yngre jernalder


Kærgård, Daugbjerg
– a Late Iron Age settlement with a workshop area

Excavations carried out in 2007‑09 on the edge of a meadow at the farm of Kærgård, about 15 km west of Viborg (figs. 1-2) revealed evidence of houses and activities dating from the Germanic Iron Age and Viking Age (c. AD 400‑950). Even though an area of about 14,000 m2 was uncovered, the site has not been fully excavated. But even so, 19 houses, 20 pithouses, 27 fences, 40 wells and waterlogged pits and four drying pits were located.

The houses
Only a few of the 19 houses will be mentioned here. House K2/K3 (fig. 5) is presumed to be a longhouse with dwelling and byre that was rebuilt once or twice on the same site. The house dates from the Early Germanic Iron Age or perhaps slightly earlier. House K8 (fig. 6) has a special extension to the north and is dated to the Late Germanic Iron Age or Early Viking Age. House K45 (fig. 7) is a smaller building and could be some kind of workshop; it is dated to the Viking Age. The rest of the houses that were fully uncovered are smaller, being either two-aisled or three-aisled workshop buildings.

Twenty pithouses have been identified even though not all of them were very well preserved. The finds from them are quite varied and include spindle whorls, a whetstone, bronze tweezers, an iron needle, an amber bead, a glass bead, an arrowhead and an iron knife (fig. 8). One pithouse was found to contain potsherds from at least 26 different hemispherical vessels; nine of these had a hole for a repair (fig. 9), indicating that this building was probably used for repairing pots.

Wells and waterlogged pits
The 40 structures can be divided into five groups: natural ponds, smaller waterlogged pits (10), wells without a lining (10), wells with a lining (13) and basins (3). The latter three groups in particular have yielded some interesting information.

Dendrochronological analysis of the wood has been carried out at Wormanium and the Danish National Museum, resulting in some cases in very precise dates.

The wells with a lining vary in construction: Four have a woven wattle lining (fig. 10‑11), two are lined with branches (fig. 12), two have planks and reused timber, two comprise hollow tree trunks (fig. 13) and a third has half a tree trunk.

The basins are rather shallow ponds, with logs laid out to walk on (fig. 14); one even has a layer of small branches at its base (fig. 15).

Some of the wells without a lining probably originally had one that was removed when the well was demolished. A ladder was found in each of two smaller wells without a lining; one had just a single step, the other had three (fig. 16).

A further type of structure should be mentioned: pits used for heating. Four of these contained heat-damaged stones and charcoal, a fifth held a large charred tree trunk, while another two were reused wells, almost completely backfilled, then lined with red-burnt clay. The purpose of these structures could have been for heating or drying.

The waterlogged conditions have resulted in excellent preservation, with numerous wooden artefacts being preserved. A small spoon, parts of a wooden bowl, small clubs and various items of unknown function have been found (fig. 18). There is also building timber, several wagon axles and an arrow-shaped ard share (figs. 19 and 20). Pieces of rope (fig. 21) were found in one well and another contained pieces of rolled birch bark (fig. 22).

Two wooden lures (fig. 23) represent quite unique finds. One is 50 cm long and made of willow wood. Its mouthpiece is very well preserved and has a binding of lime bast. The second lur is about 80 cm long and broader than the first. Only five other examples are known from Denmark: one from Herning Torv, three from Holing and one from Nydam.

Scientific analyses
Examples of animal bones from the site are shown in figure 24. Cattle are fairly dominant, but horse is also surprisingly common. The wood used for various purposes was also investigated. Figure 25 shows the species used for well linings and figure 26 the wood dropped or thrown into a pool. Ten different species have been identified. Two pollen analyses are shown in figure 27.

Analyses of plant remains from the wells were carried out to examine whether there was specialised production of textiles of nettle and/or flax. However only a few flax seeds were found and although there were fairly numerous nettle seeds, this was insufficient to prove that retting had been carried out in the wells and ponds. Neither was any evidence of other functions found (fig. 28).

Plant macro-remains from the pithouses include various cereals and weeds (figs. 30 and 31). House K45 also yielded several different cereals, mostly from the middle of the house where activities may have been concentrated.

Functions of the wells, pithouses and other structures
There seem to be too many wells just to provide drinking water, so other possible functions have been considered. The Viking Age settlement excavated at Næs on Zealand also had quite a large number of pithouses and wells, and in some of the latter were found bundles of flax stems. These wells had been used as retting pits for flax and the pithouses were small textile workshops. Only a few seeds of flax were found at Kærgård, but there were some nettle seeds. The botanical remains are consequently very sparse, but the archaeological features indicating textile production are more numerous (fig. 32). The many wells and waterlogged pits, ladders and logs giving access to the basins all indicate the presence of retting pits, and some drying pits could have been used for drying the plant stems before breaking them. Spindle whorls in the pithouses indicate that these could have been used for textile production.

The way the site is structured is also rather unusual. In the southern part there appear to be three typical farm units (fig. 33), while the concentration of pithouses in the north seems more likely to represent a workshop or production area. Smaller working units (all outside the fences) can be seen in at least three places at the site: These comprise a retting pit, clean water wells, drying pits and smaller workshop buildings (fig. 34).

Perspectives and conclusions
A workshop area like that located at Kærgård has not been found at any other site in Viborg Museum’s area. At Duehøj SV there were three wells and pithouses, but no retting or drying pits, at Højlund Spangsdal there was a drying pit and a waterlogged area but no pithouses, and at Spangsbjerg the retting pits, drying pits and pithouses were distributed among the farm buildings.

Other sites, such as Næs at Zealand and Seden Syd at Funen, show a greater similarity to Kærgård. However, both of these sites also have evidence of trade, of which there is no sign at Kærgård.

Iron production sites represent another type of specialised site. They are well known in southwest Jutland where the large numbers of iron-smelting furnaces at some sites indicate that the production was greater than for the village’s own consumption. Such sites have also been found closer to Kærgård at sites excavate by Silkeborg and Herning Museums. However, these sites also lack evidence of trade.

These specialised sites indicate that it is necessary to understand the organisation of the Late Iron Age settlement in a more complex way.

The excavation at Kærgård has revealed an agrarian settlement with a workshop area indicating that there was specialised production, probably of textiles, that was intended for trade with other settlements. The fact that trade and exchange became increasingly important during the Germanic Iron Age and Viking Age has been known for a long time, but we do not know of many specialised sites as that at Kærgård. We do not know whether they were controlled by a chieftain at the site or located further away, but these specialised sites are yet another piece in the jigsaw puzzle of Late Iron Age settlement structure.

Kamilla Fiedler Terkildsen
Viborg Museum

Marianne Høyem Andreasen
Moesgaard Museum





Terkildsen, K. F., & Andreasen, M. H. (2014). Kærgård ved Daugbjerg – Bebyggelse med værkstedsområde fra yngre jernalder. Kuml, 63(63), 65–108. https://doi.org/10.7146/kuml.v63i63.24461