Lab Summary Week 9: Turf Building

By Evan Orjala and Max Serota


Early Icelandic Viking settlers encountered an environment that lacked the materials traditionally used for housing in their native Scandinavia; Iceland’s relatively young landscape lacked large forests with wood with which to build the walls and roofs of permanent shelters, nor was its volcanic rock suitable for building sturdy walls (Byock 2001, 34). Viking Age Icelanders thus turned instead to turf–blocks of grass and the upper layer of dirt held together by roots–to construct their homes, adopting a technique previously used only for insulation in Scandinavia. They used wood and rock only minimally as the foundation and frame of their buildings, building their walls and roofs out of many different types of turf blocks. 

A picture of a reconstructed traditional Icelandic turf building.
Image of the Þjórsárdalur reconstruction of a traditional turf house, from

These same techniques were employed in the construction of temporary shelters called booths where Icelanders stayed during regional and national parliamentary assemblies. Archaeological and historical evidence indicates booths were made out of a permanent turf wall along with a potentially more temporary and portable wooden frame and a cloth roof that were only put up when the booth was actively in use. It is unclear, however, exactly what the wooden structure consisted of. One potential hypothesis is that a wooden tent, such as those found in Viking ship burials, was used as the roof of the booth. For our lab this week we thus attempted to test this hypothesis by constructing a small rectangular turf booth that could accommodate the reconstructed Viking-style tent that we built last week. (You can learn more about the tent we built last week here).

This experiment, however, was intended to do more than merely test this hypothesis. We also sought to gain an experiential understanding of the processes underlying turf construction: we attempted to understand how the properties of turf influenced and constrained the creation of turf blocks and turf buildings, how easy different types of turf blocks were to harvest and build with, and the social and economic implications of the communal activity of turf building. 


Block TypeStrengirKlömbruhnausGlaumbaejar-hnausKvíahnausHornhnaus
DefinitionTurf strips that were thin, long, and taperedWedge- shaped “Clamped blocks”“Glumbauer blocks” that run perpendicular to the wall Rectangular “Milk pen blocks” that  alternate lengthwise and crosswiseSquare corner blocks that form the corners of the booth
Dimensions12 in x 24 in x 3 in (thickest)6 in x 12 in x 3 in (thickest)6 in x 24 in x 3 in6 in x 12 in x 3 in12 in x 12 in x 3 in
Different types of turf blocks.

The class split up into three-person lab groups, with group A tasked with cutting klambra, groups B and C cutting strengir, group D cutting glaumbaejarhnaus, group E cutting kvíahnaus, and group F cutting hornhnaus. We began by laying out the footprint of the tent we constructed in Week 8 lab to get a sense of the interior area and approximate perimeter of the walls. Some groups began by cutting blocks from the floor of the interior area, while others found space in the field to begin cutting. Once the whole floor area was cut out, all groups began cutting their blocks in the field. 

The tent footprint laid out and initial interior excavation.

To cut the turf, we used a variety of modern and recreated Viking-era tools. These included spades, a garden edger, a mattock, beekeeping knives, and two experimental tvïïskeri. Harvesting a block typically involved cutting out the perimeter of the block with a spade, and then cutting out the bottom of the block from the dirt below using a beekeeping knife. Once the first block in an area was removed, leaving a hole on one side, the latter process could be sped up by separating the block from the ground using the adze end of a mattock. 

Tamping down the walls.

When blocks began to accumulate, each group (or in the case of the strengir, groups B and C) began constructing their respective wall out of their particular block type. After each layer of blocks, we tampened the turf wall by walking on it. Once we were satisfied with the height of the walls (and nearing the end of lab period) we added one final layer of strengir with grass facing upward on top of all four walls. We then reconstructed the entire tent and placed it over the booth. 

The position of different turf block types in the booth.

Data and Analysis

Turf Harvesting Data

Harvesting Kvíahnaus.

The speed with which turf could be harvested seems to have been impacted heavily by the type of turf block that a group was attempting to harvest. Strengir, for example, took far longer than either Klömbruhnaus or Glaumbaejar-hnaus. The amount of time it took to harvest turf blocks of different types is displayed in the table below. 

Block NumberStrengir (C)Strengir (B)KlömbruhnausGlaumbaejar-hnaus
4025 minutes for all blocks
Time it took to harvest blocks of different types. Credit to Emmet Forester, Glen Norvell, Julia Tassava, and Sadhana Mandala.

Notably, almost every group rapidly improved the speed with which they were able to harvest blocks. The graph below illustrates the increase in speed by different groups.

A challenge of turf harvesting was keeping the blocks consistent sizes. Almost all block types varied substantially from block to block. The variation in size of some Kvíahnaus is shown in the table below. 

Brick Length (cm)Brick Height (cm)Brick Width (cm)
Variation in turf block size. Credit to Margaret De Fer.

Final Structure Dimensions

Qualitative Observations

Working with a new and sometimes unpredictable material like turf yielded some frustration. One of the most common observations was that it was very difficult to adhere to the dimensions outlined in the lab manual. Especially at first, with no pre-dug area from which to cut inward, it was nearly impossible to create a taper on one end. The random appearance of rocks and roots complicated this process further. It appears that most groups across all block types settled into the habit of removing the blocks in crude form and refining them by hand. One student suggested that his group worked two to three times faster when not worried about adhering to dimensions. Furthermore, we supplemented the strips and blocks with extra loose dirt and turf pieces during the construction process to level out the walls and provide additional support, which eliminated the issue of improper shapes and sizes. 

Some participants also recall difficulties with the size of the strips, which were wide enough that it was difficult to get all the way underneath them with the spade. This necessitated lifting the block while another person cut the remaining dirt and roots from below, sometimes leading to the turf falling apart. Strips could also collapse during transport, although this could be prevented with the inclusion of a third carrier rather than two. Some students noted that the strips and blocks became easier to move once they had dried in the sun for a while and moisture had been removed. 

Students also reported a high variance in tool efficiency, with more modern tools like the serrated beekeeping knives and mattock outperforming the handmade and non-serrated tviïskeri. 

Many students concluded that harvesting turf and constructing turf walls was a fairly straightforward process with a low barrier to entry. That being said, this was likely the most exhausting lab (not least due to 80º+ heat), and becoming skilled enough to harvest and build efficiently would take years of practice. We were all thrilled to see the successful product of our exhausting labor. 


This lab, although tiring, ended up being a lot of fun. It was really nice to get to see our booth come together with the tent that we built last week. Indeed, the booth seemed to accommodate the tent structure well. While this isn’t exhaustive proof that this was true Viking practice, we believe that the combination of tent and booth structure constituted a spacious living arrangement with solid structure to lean up against and ample height for standing and moving around. One member of the class who has been spending time in the tent over the past few days has reported that, despite ample rainfall, the interior of the tent has remained very dry. 

One area for future research would be to experiment with different types of tent material. We used a canvas cloth, and it appears to be performing well in wet conditions, but perhaps a different material would have been more likely used by the Vikings. How might the canvas perform in winter? 

Many people also commented on the teamwork element of this lab (as opposed to spinning wool, perhaps) as a major factor in their enjoyment. While spinning wool was a very social activity, one was not necessarily reliant on the people around them to complete the task at hand. Cutting turf, on the other hand, would likely have been next to impossible without other people to cut while others are lifting or vice versa, carry the turf from the harvest site to the build site, and to provide relief when one gets tired. The physical labor inherent to turf building was made bearable (even enjoyable) and went by very quickly because of the social conditions brought about by working with and around others (especially on a task that, at times, didn’t require intense focus). 

Trimming turf walls.

The communal nature of turf building.
Erecting the tent frame on the booth.

0 thoughts on “Lab Summary Week 9: Turf Building

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.