Written by Ingeborg Sæhle, archaeologist at the Norwegian Institute for Cultural Heritage Research

In 2016, archaeological investigations in the central part of Trondheim (Søndre gate 7-11) uncovered the remains of what is believed to be the St. Clemens church purportedly built by King Olav Haraldsson in around AD 1015.

As the excavation continued in 2017 it was soon realized that the church was not originally erected on virgin soil, but rather on 1, 2 m (3, 9 ft.) of accumulated culture layers and corner-timbered buildings from the late Viking age. Due to the very early date of the first church (dated by dendrochronology and 14C to the period AD 1010-1024), the buildings beneath it were potentially some of the oldest urban corner-timbered buildings recorded in Norway.

When the excavation was finished in October of 2017, four occupational phases with related buildings had been unearthed beneath the church.

The bay

The early urban settlement in Trondheim was centered around a small bay which extended northwards in to the Nidarnes peninsula from the river Nidelva, creating a narrow spit of land between the river and the bay where buildings were arranged in organized plots (Fig. 1).

Fig. 1: Map of the Nidarnes peninsula with the discussed bay area marked in blue. Kaupmannastretet is marked as a north-south orientated dark brown line. The excavation is marked with a star. The proposed extension of the bay found in 2017 is marked as a lighter blue. (Map by Kristoffer Brink, adapted from Christophersen & Nordeide 1994: 68, fig. 42)
Fig. 1: Map of the Nidarnes peninsula with the discussed bay area marked in blue. Kaupmannastretet is marked as a north-south orientated dark brown line. The excavation is marked with a star. The proposed extension of the bay found in 2017 is marked as a lighter blue. (Map by Kristoffer Brink, adapted from Christophersen & Nordeide 1994: 68, fig. 42)

The plots were aligned towards the bay in the east and the river in the west, and a north south orientated street; Kaupmannastretet, bisected the two. The street curved around the eastern end of the bay, the northern end of which was uncovered through our excavations in 2017.

The earliest activity in the excavation area is represented by a series of organic refuse layers deposited in the bay, creating thin lenses of organic material mixed with clay and silt as the tide washed them out.

The act of throwing refuse into the bay seem to have been a deliberate act, and coupled with post-glacial rebound it caused this area to gradually dry up. This resulted in the disappearance of the bay’s northern end sometime in the 900s AD.

Oldest evidence of urban corner-timbered buildings in Norway

After the bay had dried up, the natural ground was substantially leveled up with river sand in order to create an even surface on which to build.

Much energy was invested in creating a suitable plot of land, and the activity is similar to the later construction of large-scale terraces seen further south in the bay in the 11th century (Christophersen 1994a: 141ff). After the area was leveled a property boundary was laid out, dividing the excavation area into two separate plots of land.

What separates the early urban settlement in Trondheim from contemporary urban settlements elsewhere in Norway is the early use of corner-timbering as construction method for buildings.

Elsewhere in Norway corner-timbered buildings do not appear in the urban settlements until the mid 11th century, whereas Trondheim has evidence of corner-timbered buildings from the mid- to late 10th century (Molaug 2007, Christophersen 1994: 162f, Brendalsmo & Molaug 2014: 155).

 

Plan-drawing of Building 1 and Building 2 with property boundary in the form of a ditch between them. Both buildings are truncated by younger disturbances. Drawing: Kristoffer Brink, NIKU.
Fig. 2. Plan-drawing of Building 1 and Building 2 with property boundary in the form of a ditch between them. Both buildings are truncated by younger disturbances. Drawing: Kristoffer Brink, NIKU.

This is mirrored at our site, where the first constructed building (Building 1) on the newly established plot of land was a large, corner-timbered building separated into at least two rooms ( Fig.2). On the property next to it, a building constructed in post-and-plank technique (sleppverk) was erected (Building 2, Fig. 3). The corner-timbered building had a beautiful wooden floor and wall bench (moldbenk) in the central room and an earthen floor in the northern gable room.

 

Fig. 3: Photo of post-and-plank outer wall construction in Building 2. Photo taken in section. Photo: NIKU
Fig. 3: Photo of post-and-plank outer wall construction in Building 2. Photo taken in section. Photo: NIKU

There is clear evidence that Building 1 burnt down, and that the area was subsequently leveled. However, a new building (Building 3) was quickly erected on the exact same spot as the old one. Building 3 is a three-room corner-timbered building with a wooden floor, wall bench and corner-hearth in the central room (Fig. 4).

 

Fig. 4: a: Building 1. Measures 4, 2 m in width, preserved length app. 6 m (complete length unknown). b: Building 3. Measures 4, 7 m in width, preserved length app. 6 m (complete length unknown). Photo: NIKU
Fig. 4: a: Building 1. Measures 4, 2 m in width, preserved length app. 6 m (complete length unknown). b: Building 3. Measures 4, 7 m in width, preserved length app. 6 m (complete length unknown). Photo: NIKU

Connected to this building are traces of fine metal-working in the form of crucible fragments, a fragment of a crucible tong, copper-alloy foil, and a lump of brass (Fig. 5). Traces of trading activities were also found in the form of cut silver coins and weight- related items.

Fig. 4: a: Building 1. Measures 4, 2 m in width, preserved length app. 6 m (complete length unknown). b: Building 3. Measures 4, 7 m in width, preserved length app. 6 m (complete length unknown). Photo: NIKU
Finds connected to Buildings 1 and 3, associated fine metal-working. a: Lump of copper-alloy. b: Thin flake of copper-alloy. c: Possible remains of crucible tongs. d: Crucible fragment with preserved rim. Photo: NTNU and NIKU

Building 3 has been dated by dendrochronology to 981/982 AD, which is the felling date of the timber. This means that the building was probably erected some years after this.

Building 1 beneath it was probably erected 20-30 years previous, though no secure dates have so far been possible to obtain from this building.

It is therefore possible that Trondheim started the trend of corner-timbered houses in urban areas more than 50-70 years before the rest of Norway’s other urban centers adopted the technique.

 

Intentionally destroyed

A final phase of building activity followed the destruction of Building 3 by fire, and two new buildings were erected on the plot (Buildings 4 and 5). Even though the plot boundary stayed the same, the property shifted a couple of meters further to the south, creating space for two corner-timbered buildings with a wood-paved backyard area between them (Fig. 6, fig. 7).

Fig. 6: Photo of Building 4 to the far right, and Building 5 to the far left. Building 5 was in very poor condition. Between the buildings was a wood-paved area which was not yet exposed when the photo was taken. Photo: NIKU
Fig. 6: Photo of Building 4 to the far right, and Building 5 to the far left. Building 5 was in very poor condition. Between the buildings was a wood-paved area which was not yet exposed when the photo was taken. Photo: NIKU
Fig. 7: Close-up of the north-western gable corner in Building 4. In frame is a section-view of the corner, showing two layers of logs (omfar) in the northern gable-wall, and the sill-beam in the western long wall. The larger photo shows the corner after the removal of the upper log in the northern gable wall. The sill-beam in the western long wall is shown on top of the sill-beam in the northern gable wall. Photo: NIKU
Fig. 7: Close-up of the north-western gable corner in Building 4. In frame is a section-view of the corner, showing two layers of logs (omfar) in the northern gable-wall, and the sill-beam in the western long wall. The larger photo shows the corner after the removal of the upper log in the northern gable wall. The sill-beam in the western long wall is shown on top of the sill-beam in the northern gable wall. Photo: NIKU

Whereas the earliest buildings were associated with trade and metalworking, these new buildings appear to have had more domestic functions. This is reflected in the in-situ remains of fishing weights in the backyard area, and loom-weights along the wall inside Building 5 (Fig. 8, fig. 9).

 

Fig. 8: Archaeologist Kristoffer Grini excavating the area between Buildings 4 and 5. Situated directly outside the gable wall of Building 4 were several net-sinkers, possibly from a fishing-net hung here and collapsed in-situ. Photo: NIKU
Fig. 8: Archaeologist Kristoffer Grini excavating the area between Buildings 4 and 5. Situated directly outside the gable wall of Building 4 were several net-sinkers, possibly from a fishing-net hung here and collapsed in-situ. Photo: NIKU

 

Fig. 9: Possible loom-weights and spindle-whorls found inside Building 5, along the southern gable wall. Photo: NIKU
Fig. 9: Possible loom-weights and spindle-whorls found inside Building 5, along the southern gable wall. Photo: NIKU

There is so far no evidence that these buildings burnt down, but there is evidence of deliberate destruction. A massive refuse layer sealed the buildings, on top of which a layer of burning could be observed.

It appears that the buildings were intentionally destroyed and filled in, and it is on top of this extensive layer of infill and subsequent burning that the church was built. We know that the church was built in the first quarter of the 11th century, which means that the destruction of the underlying buildings must have taken place sometime before this.

 

An urban settlement before the Viking king Olav Tryggvasson?

The creation of large sand terraces all along the bay shows a massive investment in the creation of new land, and the establishment of fixed property boundaries prior to the construction of the first buildings indicates the presence of a preconceived town plan. Both these factors are indicators of a centralized power-structure, and could therefore be related to the establishment of the settlement by royal design.

According to the later sagas Trondheim (Nidaros) was founded by King Olav Tryggvasson around AD 997. However, archaeological investigations within the city have revealed substantial settlement traces pre-dating this proposed founding date for the city.

Further south in the bay another corner-timbered building (K11) has been dated to the mid- to late 10th century (Reed TA 1991/2: 21), and this building, coupled with the very early dates for our buildings, indicate that Trondheim was in fact established as an urban settlement several decades prior to Olav Tryggvasson’s royal urban center of AD 997.

he early dates for the initial urban settlement are very interesting viewed in light of the contemporary political situation in Norway – could it in fact have been one of the Earls of Lade who initiated the urban settlement in Trondheim? The new dates raise many questions regarding which central power had control over the area.

An equally interesting aspect is the later destruction of the pre-existing profane buildings in order to build a church on the grounds in the beginning of the 11th century. The buildings appear to be intentionally destroyed and leveled in order to create a new plot of land for the church, which tells us that the buildings were probably destroyed by the command of the king.

This raises many questions regarding why precisely this spot was chosen for the church, and who the people were who occupied this area before the church was built. This excavation has provided many new insights into the early urban settlement in Trondheim, but it has raised equally many questions regarding the character, timeline and function of this settlement.

An extraordinary outcome of the excavation are the remains of the earliest preserved corner-timbered buildings so far found in an urban context in Norway, which again raises many questions regarding how this technique was brought here, from where, and why exactly Trondheim was the chosen area for the initial introduction.

 

Litterature

Brendalsmo, J., & P. Molaug. 2014. To norske byer i middelalderen – Oslo og Tønsberg før ca. 1300. Collegium Medievale, vol. 27. 136-202

Christophersen, A. 1994a. Gård og grunn. In: (eds.) A. Christophersen & S. W. Nordeide. Kaupangen ved Nidelva. 1000 års byhistorie belyst gjennom de arkeologiske undersøkelsene på Folkebibliotekstomten i Trondheim 1973-1985. Riksantikvarens skrifter, vol. 7. 113-212

Christophersen, A. 1994b. Strete, havn og kirkegård. In: (eds.) A. Christophersen & S. W. Nordeide. Kaupangen ved Nidelva. 1000 års byhistorie belyst gjennom de arkeologiske undersøkelsene på Folkebibliotekstomten i Trondheim 1973-1985. Riksantikvarens skrifter, vol. 7. 67-112

Molaug, P. B. 2007. Occupying town plots. On the early development of habitation in Norwegian medieval towns.  Medieval Europe Paris 2007, 4th International Congress of Medieval and Modern Archaeology.

Pilø, L. 2007. The settlement: Character, structures and features. In: (eds.) Dagfinn Skre, Kaupang in Skirringssal. Kaupang excavation project publication series, vol. 1. Norske Oldfunn XXII. Aarhus University Press. 191-222

Reed, I. 1991. TA 1991/2. Norges bank. Unpublished excavation report.  Riksantikvarens utgravningskontor i Trondheim.

Tesch, S. 2001. Houses, town yards and town planning in late Viking Age and Medieval Sigtuna, Sweden. In: (eds.) Manfred Gläser, Lübecker Kolloquium zur Stadarchäologie 3: Der Hausbau. Lubeck. 723-741.