Mellemtoft Øst – En tidlig middelalderlig landbebyggelse nær Tvis Kloster


  • Bo Steen


Mellemtoft Øst, tidlig middelalder, landbebyggelse, Tvis Kloster


Mellemtoft Øst
An early medieval rural settlement near Tvis Kloster

In 1999, Holstebro Museum in vestigated a rural settlement from the Middle Ages in Mejrup parish three kilometers east of Holstebro (fig. 1-2). During the investigation, a system of squares and ditches taking up an area of 6000 square meters were unearthed.

The medieval structures comprises a road track running north-south between two parallel croft ditches,and east of this, three longhouses, a pen, and five Dutch barns. Moreover, a Bronze Age house site and a few pits and postholes from the Neolithic Age and Bronze Age were uncovered.

House I had a length of 30.5 meters and an east-west orientation.There were traces of seven roof-bearing posts placed between 4 and 6 meters apart along the axis of the house (fig. 3). At both gables, the outermost roof-bearing posts were placed in the wall. The southern house wall consisted of an inner and an outer row of posts, placed 0.75-1.1 meters apart. The width of the house, measured between the inner walls, was 5 meters.The northern house wall seemed to have had only one row of posts. However, some of the scattered postholes were situated at such a distance from the axis of the house that they might indicate that part of this wall also had two rows of posts.

The inner southern house wall had three entrances, each 1.25-meters wide and moved a bit into the house. One side of each entrance had been placed next to a roof bearing post, which indicates that the hou se had more rooms. Postholes on either side of the roof bearing post next to the westernmost entrance were probably from a partition wall. The postholes from the wall posts contained no traces from the posts, and it is impossible to determine whether the posts of the outer row were upright or slanting.

House II was situated 30 meters south of house I and parallel with it. It had a length of 26 meters and a width of 4.5 meters, measured between the inner walls (fig. 4).

House III was represented by a g roup·of postholes probably representing several phases of one or more houses and covering an area of 18X6 meters in the north western part of the excavation, next to ditch C (fig. 5).

In the north western corner of house I, an approximately 9x9-meter large square was not contemporary with the house. The south western corner of the structure had a 5-meter wide opening. This structure is interpreted as a pen.

Several postholes very close to the eastern gable of house I and the western gable of house II did not seem to belong to the house constructions, but probably represent detached buildings. From these postholes,it was possible to isolate a Dutch barn by house I and three such barns next to house II. However, the rest of the postholes outside the houses indicate the existence of still more structures. West of the pen, another group of postholes could be interpreted as a Dutch barn, whereas other postholes gave no clue as to their function.

West of the longhouses, croft ditch A ran north-south, and ditch B ran parallel to th is at a distance of 9-14 meters. Between these ditches was a road track (fig. 2,6-7). The ditches and the road continued beyond the edge of the excavation.

North of the presumed house III, another ditch ran at an angle from ditch A. To the east, ditch C ended close to a modern ditch, but it may continue in to the not investigated area to the east. South of the corner, at the junction between ditch A and ditch C, there was a possible entrance, as at this point, ditch A was not detectable across a 3.5-meter wide stretch.

The fact that the road and the croft ditches A and B were parallel indicates that they were contemporary. Possibly, not only the road, but also the area between the ditches functioned as a common cattle walk.

A common feature shared by the medieval structures was a general lack of finds.The artifacts only amount to a few potsherds from the early 13th century, fragments of a grinding stone of Rhine basalt and fragments of medieval bricks. Samples for a C-14 dating were taken from the postho es of two roofbearing posts and an entrance post. A coin found in the excavated area south of house II dates back to 1234-1259.

No fireplaces or traces from floor layers were found in any of the buildings at Mellemtoft Øst. However, modern ploughing could have gradually destroyed these. In the assumed house III , the postholes contained potsherds, medieval bricks, fragments of grinding stones, and fire-cracked stones. The finds may indicate that house III was a dwelling house.

Contrary to house III, the two almost identical houses I and II with central posts contained almost no finds.These houses were situated in the same area as the Dutch barns and the pen, and their function may therefore be interpreted as outbuildings rather than as dwellings.

House I and II at Mellemtoft Øst are interpreted as houses with straight walls and a central row of roof bearing posts combined with a “lean-to” – an unusual type house for medieval Denmark. Houses of a type similar to I and II usually date from the 12th century. The settlement is thought to have existed from the mid-12th century until the mid-13th century.

Ditch A, is disturbed by neither the long houses 1-111, northe pen, nor the Dutch barns, and it is interpreted as the western demarcation of the croft to which the buildings seem to belong.

The parallel croft ditches,A and B, and the road between them, were parallel and had the same structure. They are therefore interpreted as contemporary. If ditch C is interpreted as a dem arcation of the partly uncovered farm site, then another croft must have been situated north of it. Simil rly, dit h B may be int rpreted as the eastern demarcatio  of a third croft, which may have been situated west of the road. All these farms must have used the same road and cattle walk. Alternatively, the partly uncovered farm site was the only one. However, if this is the case, we are left wondering why ditches were necessary on both sides of the road.

It is an obvious thought that the settlement at Mellemtoft Øst was somehow connected with the contemporary monastery of Tvis Kloster, a mere kilometer to the Southeast.

The monastery was founded by Prince Buris, the son of Henrik Skadelaar, whose father was a brother of King Erik I the Evergood. Prince Buris owned landed property around Holstebro, and in a deed signed on the 20th of March, 1163, by King Valdemar I the Great, and others, he gave part of his estate to the Cistercian Order, which was to establish a monastery where the farm of Tvisel was situated. Apart from Tvisel, the monastery had other possessions, but none of them lie in Mejrup parish, and so the settlement at Mellemtoft Øst probably did not belong to the monastery. However, the two contemporary settlements must have known each other, and some connection is suggested by the road, which may well have led from the settlement to the monastery. None of the roads on the maps from the early 19th century – which are the oldest maps of this area – can be identified as the one excavated in the settlement. Whether or not the road led to the monastery will be difficult to prove archaeologically, as the lake connected to the hydroelectric power station has flooded the areas on both sides of the old river.

Bo Steen
Holstebro Museum

Translated by
Annette Lerche Trolle





Steen, B. (2002). Mellemtoft Øst – En tidlig middelalderlig landbebyggelse nær Tvis Kloster. Kuml, 51(51), 267–283. Hentet fra