The Frontier of Archaeological Reconstruction

Horse Sacrifice at Eketorp Fort, Sweden

By: Jan Olofsson and Egil Josefson

Originally Published in 2007

View PDF

Eketorp Fort Diagram
“The Development of Eketorp Fort”

In 1978 the archaeological remains of Eketorp Fort were reconstructed on Öland Island in the Baltic Sea, establishing the first Swedish living history museum to combine an archaeological reconstruction with an experimental archaeology and popular education program. The mission of Eketorp Fort is to bring the Swedish Iron Age (500 BC–AD 1050) and Middle Ages (AD 1050–1526) to life using reconstructions, guided tours, school programs, interpretive role-playing, experimental research, traditional handicrafts, and interactive activities—all led by seasonally employed archaeologists. Our goal is to help the 50,000 people who visit us each season understand the conditions under which people lived in the past by allowing and encouraging visitors to watch and participate in characteristic activities from the periods we have recreated.

Eketorp Fort is normally no great exception from other living history museums. Our visitor activities are generally uncontroversial and appropriate for children. For some of us, however, recreating the past poses some interesting questions. What sort of “past” are we actually reconstructing? Is it a mass-produced, cute and sanitized, “lite version” of the Iron Age and Middle Ages? Or does it come as close as possible to our latest understanding of what these periods were like, even if this means recreating aspects of the past that might be controversial or offensive? What are the limits to what can be reconstructed? In 2004 and 2005 we decided to probe them by illustrating what an Iron Age sacrificial site might have looked like.

Horse Sacrifice At Eketorp

During the excavations of Eketorp in the 1960s and 70s, the remains of a shallow lake were located outside the east entrance of the fort. Although the lake had been drained in recent times, a small water hole still exists just outside the reconstructed fort’s curtain wall. Excavations in this area revealed a large number of hazel rods and animal bones from cattle, goats, sheep, and several horses. The concentration of horse crania (heads) and leg bones was especially striking.

Numerous other Iron Age sites in Scandinavia have revealed similar evidence of apparent horse sacrifice—the means by which prehistoric people communicated with their gods. For example, based on the discovery of about 100 horses in the Skedemosse Bog, it has been identified as the central sacrificial place on Öland Island during the Iron Age. Given the finds at Eketorp, the lake outside the fort was probably used as a sacrificial site.

In 2004 we decided to reconstruct this aspect of the site to illustrate Iron Age religious beliefs for our visitors. The first issue that arose was how “real” should our reconstruction be. Some argued that our visitors would never accept the use of real horse remains—too many visitors would take offense and it would be too disgusting for children. Others emphasized the need to be accurate and the important role our reconstructions played in bringing the past to life, even if it meant using dead horses. In the end, we based our reconstruction on archaeological reports, early medieval historical sources, and anthropological studies about horse-using societies to identify a likely sacrificial configuration for our preliminary reconstruction of 2004 and our full one in 2005.

The Evidence For Horse Sacrifices

No one can say for certain what Eketorp’s sacrificial site once looked like. What we can do, however, is try to base our reconstruction on as many of the available sources of information as possible. The best place to start is within Scandinavia, where several similar sacrificial sites have been found to contain only certain horse parts—the heads, tails, and feet.

What did the ancient Scandinavians do with these parts? One historical account from AD 950 gives us some guidance. While visiting the Scandinavian market town of Hedeby (in modern-day north Germany), the Arab tradesman al-Tartuschi witnessed villagers placing animal heads on poles to honor their gods during a celebration.

Was this a common practice? It appears so, and one that was widespread across northern Eurasia for a long period of time. Similar practices have been recorded by anthropologists working in Central Asia where, among the Altai people, horses are often sacrificed and then skinned in a way that leaves the head, hoofs, and hide as a single piece. This hide, with its extremities, is then placed on a long pole, using the tip of the pole to support the back of the horse cranium.

This image of a horse head mounted on a pole is also recognizable in the medieval Icelandic saga of Egil Skallagrimsson, which dates from the 10th century AD (although it was first recorded in writing during the early 13th century). In one episode Egil erects a nidstång (“a pole of scorn”), consisting of a horse head mounted on a hazel pole, which he then points in the direction of his enemies to taunt them. This example is particularly interesting considering the discovery at Eketorp of numerous hazel rods in the same context as the horse crania.

Did Eketorp’s sacrificial site use hazel rods to mount horse heads? Unfortunately, any evidence of postholes at Eketorp has long since disappeared with the dredging of the waterhole since the Iron Age. However, other comparable Swedish archaeological sites (e.g. Hjärup and Fredriksberg) do provide evidence of such postholes.

Our Reconstruction

With the evidence available we set out to reconstruct a reasonable facsimile of a horse sacrifice, using horse carcasses obtained from a slaughterhouse. Our reconstruction consisted of a sacrificial site focused around the upright remains of two horses—the offering for the present year—surrounded by the skeletal remains of numerous other horses, representing the offerings of previous years. We began by erecting diagonal hazel poles in the ground. Next we mounted the horse heads on the ends of the highest poles and inserted the horse tails into their mouths. This reconstruction was based on the discovery of horse tailbones with horse crania at the site of Rislev Mose in Denmark. We then stretched out the horse hides, with the hoofs still attached, behind the heads and draped them over a long trailing pole.

Turning to the past season’s offerings, we placed skeletal horse bones directly in or near the waterhole and around the mounted horses, situating the cranium in the center and foot bones surrounding it. This pattern was based on archaeological sites in South Scandinavia (e.g. Sorte Muld and Rislev Mose) and even some far to the east in the Caucasus Mountains between the Caspian and Black Seas (e.g. BajtalTschapkan). We also used pigs’ feet and heads to represent the remains of other animal bones found during the Eketorp excavations.

To delimit the border of the sacrificial site, we placed hazel rods around it. In the saga of Egil Skallagrimsson, hazel rods were positioned around the area where the leaders met to make important decisions on special occasions. Since it is clear that Eketorp had been a gathering place for centuries, this seemed reasonable. We also laid a paving of stones to the water’s edge leading to a 2 m long oak board that served as a footbridge above the water’s surface like the one recovered at Frösvi in Sweden.

Finally, to reconstruct the broader experience of sacrificial behavior at the site, we placed food offerings—bread, beer, and sausages—in pottery around the paving and the footbridge. One such offering was intended to recreate a specific type of food sacrifice identified by Anne Carlie at the site of Lake Käringsjön in southwestern Sweden. She had found ceramic pots with holes in their bottoms that contained the residue of a high-fat porridge. By way of an experiment, we poured a similar porridge into a ceramic pot with a hole in its bottom and showed that if such a vessel is placed on a few stones, it looks as if the wetland slowly sucks the porridge out of the pot—a particularly evocative demonstration for the public.

The Public Reaction

Our reconstruction began in June 2005 and lasted for about two months. The purpose was not to shock the visitors but rather to show them a part of Iron Age society which is rarely illustrated. Working actively with the reconstruction, we made it an integrated part of our ambitious archaeological education program at Eketorp. Under the guidance of experienced archaeologists, visitors were able to participate in an excavation of the sacrificial site, see a reconstruction of what it might have looked like, watch and take part in reenacted sacrifices, and learn about the site and similar ones on our guided tour.

Visitor’s questions were common and almost always open-minded and straightforward. Why have you done this? How can you support this interpretation? Are the horse heads real?

Very few visitors condemned the reconstruction—maybe one or two people out of 50,000. Sure, it looks a little horrible, but when we explained why we did it, most visitors agreed that it is important to reconstruct prehistoric times as realistically as possible. Otherwise, it is difficult to argue that you are making history come alive. In fact, visitors often demand the “truth” (as far as archaeology can deliver it) and realism as well. Is that unreasonable? Maybe living history museums should try to be more realistic.

We did, however, make one big mistake. We did not realize that we had to apply for permission from the Swedish Board of Agriculture to use the horse remains we had obtained from a slaughterhouse. Under a European Union decree concerning the handling of animal byproducts, such a failure is a felony. As a result, when this violation was realized one of us (Olofsson), the manager of Eketorp, became a potential felon. We are not hiding this fact or downplaying the importance of obtaining such permission. We should have done so, and when it became evident that Eketorp Fort should have applied for permission, we contacted the Swedish Board of Agriculture and immediately removed the pig and horse heads, which were duly destroyed.

It did not take long for the media to react. Several articles soon appeared in local, regional, and national newspapers about the “illegal horse carcasses by the fort!” The news was also sent to TT, the Swedish News Agency, and published on Teletext. Two of Sweden’s largest evening newspapers, Expressen and Kvällsposten, published articles on November 14, 2005, with the headline: “Severed Horse Heads Were Shown to Children.”

At present, it is still unclear what will happen with the prosecution of the case. Olofsson has been interrogated by the police, and the felony accusation remains.

Why Reconstruct Real Horse Sacrifices?

In her study of over 300 archaeological reconstructions in Scandinavia, Bodil Petersson discusses the expectations of visitors to museums when it comes to authenticity in reconstructions. She notes that these expectations vary depending on a particular museum’s agenda. For example, if a museum’s goal is to make a building and its prehistoric context come alive by conveying a sense of place, then visitors do not mind if the reconstructed buildings use modern materials in their construction so long as they are hidden from view. In contrast, if the museum’s objective is to show the prehistoric building process using prehistoric methods, then the use of any modern materials is considered inappropriate—even if these materials were used simply to keep costs down. Although museum visitors understand the demands of financial constraints, cutting corners by using modern materials or construction practices defeats the purpose of such reconstructions.

At Eketorp, we actually had the opposite situation—it would have been more expensive to have made the replicas of horses. That, however, was not the main reason we chose to use real horses. Besides creating a sense of a real sacrificial place by using all authentic materials, we also designed our reconstruction so it could help us understand how sacrificial sites were transformed over time into archaeological sites. In other words, by recreating authentic sacrifices year after year and letting their remains build up over time, we hoped to experimentally recreate the process by which the archaeological evidence for such sites is created. At some future date, we could have excavated our modern sacrificial site to see how it differed from archaeological examples. This would allow us to see where our reconstruction deviated from real sacrificial sites and point the way toward more accurate reconstructions of the past. We believe that if authentic materials are available and cause no harm to the understanding and experience of the reconstruction, then such materials (or methods) should be used.

The Need For Authentic Reconstructions

It is very important that we depict as many aspects of the past as accurately as possible in our reconstructions. Studies of museum displays and presentations, such as Petersson’s, consistently indicate that museums usually do not educate their visitors about the ritual, symbolic, ideological, and ethical aspects of the past. Instead, they tend to emphasize artifacts, technology, and traditional life styles and gender roles—an easily consumed and simplified vision of the past.

We maintain that museums should be courageous and challenge the preconceptions of their visitors. Rather than focus solely on everyday life, family, and security, as most Iron Age living history and open-air museums and reconstructions in Scandinavia do, we should also reconstruct ritual perspectives and socio-political features, such as warfare.

Museums should inspire questions and debate among their visitors, and when visitors ask questions, we must be prepared to explain our reconstructions and reenactments—why we illustrated something in a certain way, which archaeological evidence we based our reconstruction on, and how we interpreted it. We must also be very precise when it comes to those areas where our interpretations are uncertain.

We hope this discussion initiates a debate within and among museums (whether traditional, open-air, or living history museums), reenactment groups, and archaeological institutions about what should be depicted. If we choose not to work with as many aspects of the past as possible, we run the risk of stagnation—where all we show is a simplified, conventional picture of the past. This, in turn, will lead to the loss of public interest in visiting open-air and living history museums, consigning the past once again to a lifeless world beyond public awareness.

For Further Reading
Backe, M., B. Edgren, and F. Herschend. “Bones Thrown into a Water-Hole.” In Sources and Resources: Studies in Honour of Arrhenius, edited by G. Arwidsson, A. M. Hansson, L. Holmqvist-Olausson, B. M. Johansson, M. Klockhoff, K. Lidén, and H. Å. Nordström, pp. 327-42. Rixensart, Belgium: PACT, 1993.Carlie, A. “Käringsjön: A Fertility Sacrificial Site from the Late Roman Iron Age in South-west Sweden.” Current Swedish Archaeology 6 (1998):17-37.Görman, M. “Influences from the Huns on Scandinavian Sacrificial Customs during 300–500 AD.” In The Problem of Ritual: Based on Papers Read at the Symposium on Religious Rites Held at Åbo, Finland, on the 13th-16th of August 1991, edited by T. Ahlbäck, pp. 277-98. Åbo, Finland: Donner Institute for Research in Religious and Cultural History, 1993.

Hagberg, U-E. The Archaeology of Skedemosse II. Stockholm: KVVA, 1967.

Josefson, E., and J. Olofsson. “To Reconstruct a Sacrificial Site: Aspects of the Iron Age Sacrificial Site at Eketorp Fort, Sweden.” EuroREA: Journal of (Re)construction and Experiment in Archaeology 3 (2006):72-77.

Petersson, B. Föreställningar om det förflutna. Arkeologi och rekonstruktion. Lund, Sweden: Nordic Academic Press, 2003.

Cite This Article

Olofsson, Jan and Josefson, Egil. "The Frontier of Archaeological Reconstruction." Expedition Magazine 49, no. 1 (March, 2007): -. Accessed February 15, 2025. https://www.penn.museum/sites/expedition/the-frontier-of-archaeological-reconstruction/


This digitized article is presented here as a historical reference and may not reflect the current views of the Penn Museum.

Report problems and issues to digitalmedia@pennmuseum.org.