The preparation for the reconstruction of an iconic belt buckle from the Haithabu settlement, arguably the most elaborate example from the site, began in October 2024. During this preliminary phase, the research team analyzed the buckle’s construction and its parallels, which led to the hypothesis that an unusual strap-end with a ring was originally used with it.
The primary function of this type of strap-end is realized when its ring is slipped over the buckle’s prong and secured in a designated groove. This mechanism allows the belt’s length to precisely conform to the wearer’s waist circumference. Subsequent research within the archaeological material from Haithabu successfully identified a strap-end of this specific type. Crucially, both components were reconstructed and, for the first time, paired together. Their potential association had previously been disregarded by both academic researchers and the reenactment community.
Although both finds originate from geographically proximate locations – separated by a distance of, at most, only a few hundred meters – it is impossible to state with absolute certainty whether they originally constituted a single set or whether they represent the remains of two analogically constructed assemblages. This article is dedicated to early medieval scholars, archaeologists, artisans, and the historical reenactment community.
Map 1: Position of Haithabu on the map of Europe.
Finds
The buckle described below was discovered in 1939 during archaeological excavations conducted by H. Salmo within the so-called Semicircular Wall at Haithabu (Arents – Eisenschmidt 2010a: 38). It most likely represents a find from a disturbed grave context. The artifact is composed of a D-shaped frame and an attached buckle plate. The frame (measuring 3.5 × 2.2 cm, with a maximum thickness of 0.4 cm) features a robust prong equipped with a groove. Its surface is decorated with two triangular fields filled with relief ornaments. The exterior perimeter of the frame is further adorned with a decoration consisting of oblique lines. The rectangular plate, measuring 2.5 × 2.4 cm, is divided into several fields by horizontal lines. From these fields emerge two symmetrically oriented panels filled with a relief knotwork ornament. The open end of the plate is shaped into alternating points and curves. The buckle is made out of copper alloy and is gilded. One source specifically notes the use of niello in the ornamental fields (Arents – Eisenschmidt 2010a: 139). The strap was secured by two copper alloy rivets, and the presumed thickness of the leather strap is approximately 0.25 cm (Arents – Eisenschmidt 2010b: 92; Capelle 1968: 108, Taf. 25.6; Wamers 1985: 109).
Based on stylistic parallels, the buckle is classified as an Irish import (Wamers 1985: 32), a conclusion supported by comparisons with related finds (Arents – Eisenschmidt 2010b: 276). However, scholarly consensus on the artifact’s chronological placement is lacking. While some studies date the buckle to the 8th – 9th century based on analogies in Irish art (Capelle 1968: 76), other sources compare it with analogical buckles from Norwegian graves, suggesting a later date, likely the 10th century (Arents – Eisenschmidt 2010a: 139).
Fig. 1: Buckle from Haithabu.
Source: Arents – Eisenschmidt 2010b: Taf. 30.2.
A strap-end with a maximum length of 2.8 cm was recovered in 1963 during archaeological excavations at the settlement of Haithabu (Capelle 1970: 12). The artifact is fabricated from a copper alloy. The piece features a triangular plate, the edges of which are ornamented with a distinct zigzag line. This plate is equipped with a notch designed for securing the strap and a single rivet hole. A large, undecorated ring extends outwards from the plate. Strap-ends analogical to this specific type, which are fastened via a ring to the buckle’s prong, are known across a remarkably broad chronological and geographical spectrum. Their appearance spans from at least the Roman period (Raddatz 1957: Taf. 13-14) up to the High Middle Ages (Otroščenko – Rassamakin 1986: Рис. 8.1,9). Within the Early Medieval period, comparable sets are attested at sites such as Islandbridge in Dublin, Ireland (Bøe 1940: 43; Harrison – Ó Floinn 2014: 160-1), Kalvatn, Norway (Wamers 1985: Taf. 28.1), and Ludzas Odukalns, Latvia (Radiņš 2009: 115). A characteristic feature of buckles utilizing this configuration is that they remain slightly erect when fastened. Furthermore, the corresponding strap-ends occasionally exhibit a minor curvature.
Fig. 2: Strap-end from Haithabu.
Source: Capelle 1970: Abb. 4.1.
Fig. 3: Belt set from Kalvatn, Norway.
Source: Wamers 1985: Taf. 28.1.
Fig. 4: Belt set from Islandbridge, Ireland.
Source: Harrison – Ó Floinn 2014: 160.
Fig. 5: Belt from Ludzas Odukalns, Latvia.
Source: Radiņš 2009: 115.
Fig. 6: Belt from Chingul Mound, Ukraine.
Source: Otroščenko – Rassamakin 1986: Рис. 8.1,9.
Reconstruction
The reconstruction of both belt components was kindly executed by Silvia Stork of Ilvis Handwerk. The fabrication process began with Silvia carving the three main parts of the buckle (the buckle place, the prong and the frame) from a wax block. These models were subsequently sent to a foundry. While the prong and the frame could be cast using the sand-casting method, the buckle plate required the lost-wax technique due to the plastic relief of its decoration and resulting undercuts.
Silvia also carved the wax model for the strap-end, which she subsequently cast herself at the forge within the Oerlinghausen archaeological open-air museum in Germany. Following casting, all pieces were meticulously cleaned, smoothed, and assembled. In adherence to the original artifact, the buckle was then gilded.
Fig. 7: Wax models.
Author: Silvia Stork, Ilvis Handwerk.
During the reconstruction of the buckle’s decoration, certain necessary compromises were made. The original carved patterns were presumably filled with niello; however, due to health risks and the complexity of the traditional process, we opted for synthetic resin as a substitute. The strap-end was fashioned solely from bronze (matching the original find) and subsequently adorned with the original pattern via engraving. To ensure optimal functionality, Silvia slightly deformed (bent) the ring of the strap-end to better fit the groove of the buckle prong.
Fig. 8: Production of castings.
Author: Silvia Stork, Ilvis Handwerk.
The team of authors agreed upon using a simple leather strap for the belt. A colleague from the reenactment group RCFK undertook the preparation of the leather and performed the riveting of the metal components to the strap. The reconstructed belt was first tested during the Burgmannentage historical festival in Vechta, Germany. The buckle and strap-end proved to be highly functional. Fastening is achieved by simply drawing the strap-end through the buckle. To unlock the two parts, one merely needs to apply slight pressure on the strap-end toward the buckle plate, causing the mechanism to self-disengage. The finished set is highly successful, and we are deeply grateful to Silvia and our fellow reenactors for their contribution to its manufacture.
The following illustration displays the author’s reconstruction of the clothing and equipment based on material culture discovered at Haithabu (e.g., tunic, see Hägg 1984: 43; head covering, see Hägg 1991: 55-8; shield boss, see Vlasatý 2023). In addition to the belt, the insular influence is further highlighted by the inclusion of a thistle-brooch recovered from grave no. 312. This brooch, dated to the 10th century (Arents – Eisenschmidt 2010a: 102–3; 2010b: Taf. 29), is fabricated from a copper alloy and coated with a white metal (tin in our reconstruction). The brooch was reconstructed by Stephan Burkhardt (Zanari group). Collectively, these objects form a compelling assembly, offering an insightful glimpse into the diversity of the finds from Haithabu.
Fig. 9: Completed reconstruction of the belt.
Author’s archive.
We hope you liked reading this article. If you have any question or remark, please contact us or leave a comment below. If you want to learn more and support our work, please, fund our project on Patreon, Buymeacoffee, Revolut or Paypal.
Bibliography
Arents, Ute – Eisenschmidt, Silke (2010a). Die Gräber von Haithabu. Die Ausgrabungen in Haithabu 15. Band I (Text, Literatur), Neumünster.
Arents, Ute – Eisenschmidt, Silke (2010b). Die Gräber von Haithabu. Die Ausgrabungen in Haithabu 15. Band II (Katalog, Listen, Tafeln und Beilagen), Neumünster.
Bøe, Johannes (1940). Viking Antiquities in Great Britain and Ireland. Part III: Norse Antiquities in Ireland, Oslo.
Capelle, Torsten (1968). Der Metallschmuck von Haithabu: Studien zur wikingischen Metallkunst. Die Ausgrabungen in Haithabu 5, Neumünster.
Capelle, Torsten (1970). Metallschmuck und Gußformen aus Haithabu (Ausgrabung 1963-1964). In: Schietzel, Kurt et al. (eds.). Das Archäologische Fundmaterial I der Ausgrabung Haithabu 1963-1964. Berichte über die Ausgrabungen in Haithabu, Bericht 4, Neumünster, 9-23.
Harrison, S. H. – Ó Floinn, Raghnall (2014). Viking Graves and Grave-Goods in Ireland. Medieval Dublin Excavations 1962-81, Series B, Dublin.
Hägg, Inga (1984). Die Textilfunde aus dem Hafen von Haithabu. Berichte über die Ausgrabungen in Haithabu, Bericht 20, Neumünster.
Hägg, Inga (1991). Die Textilfunde aus der Siedlung und aus den Gräbern von Haithabu: Beschreibung und Gliederung. Berichte über die Ausgrabungen in Haithabu, Bericht 29, Neumünster.
Otroščenko – Rassamakin 1986 = Отрощенко, В. В. – Рассамакін, Ю. Я. (1986). Половецький комплекс Чингульського кургану // Археологія 53, 14-36.
Raddatz, Klaus (1957). Der Thorsberger Moorfund. Gürtelteile und Körperschmuck, Neumünster.
Radiņš, Arnis (2009). Lettlands viele Völker. Archäologie der Eisenzeit von Christi Geburt bis zum Jahr 1200, Zossen.
Vlasatý, Tomáš (2023). Decorated shield boss from Haithabu. In: Project Forlǫg : Reenactment and science [online]. [2025-10-18]. Available at: https://sagy.vikingove.cz/en/decorated-shield-boss-from-haithabu/.
Wamers, Egon (1985). Insularer Metallschmuck in wikingerzeitlichen Grabern Nordeuropas. Untersuchungen zur skandinavischen Westexpansion, Neumünster.