Rethinking the wallet from Gokstad

The wallet from the rich grave of Gokstad, Norway (early 10th century), is one of the most frequently reconstructed objects of the Early Middle Ages and is sold by many dozens of manufacturers around the world. However, its proposed reconstruction repeatedly attracts the attention of reenactors due to its ambiguity. In this article, we will take a closer look at the wallet and we will propose a new interpretation that has not yet been implemented.

Current interpretation of the wallet

The current interpretation, which is used by reenactors and which is presented in the academic literature, was proposed by Birgit Heyerdahl-Larsen in her article Gokstadhøvdingens pung (The Gokstad chiefstain´s pouch) in 1981. In the article, the author literally states:

[…] some ready-cut pieces of leather or hide which Nicolaysen believed to be the remains of a pouch, were found in the burial chamber along with the chieftain´s personal effects. Unfortunately the pieces no longer exist. Probably the leather dried up and disintergrated when no longer protected by the blue clay of the mound. A hundred years ago modern methods of preservation were of course unknown. Nevertheless I venture to make the «pouch» the theme of an article basing my assumptions on drawings of three of the leather pieces in the publication «The Viking Ship Discovered at Gokstad near Sandefjord» and a description in the main catalogue of the Museum of Antiquities which reads as follows

‚Two oval pieces of double thin hide, 13.5 × 8.5 cm, cut straight at one edge. The pieces correspond and have been sewn together at the edges. One piece has fine openwork patterning and was lined with a coloured material which must have offered a pleasing contrast to the hide at the aperture. There was preasumbly yet another patterned piece. There were also straps and an oblong piece, worn thin at both ends, decorated with three rows of closely-punched holes down the centre. The seams along the edges had left holes in all the pieces.’ “ (Heyerdahl-Larsen 1981: 104)

The text indicates that the author never saw the wallet and its interpretation based on a drawing published Nicolaysen (1882: Pl. IX.3-5) and a description that can be found today in the central catalog Unimus (Unimus 2020). The basis of her interpretation is Nicolaysen’s drawing of an oval object, to which she attaches a perforated leather belt in various ways. The basis of the two versions attached below is a two-part wallet sewn on the edges – both halves are decorated with a openwork and are lined with a textile inside and equipped with two pockets. According to the author, this base has a size of 13.5 × 8.5 cm. The version on the left places a perforated leather strip at the mouth of the pouch, while the version on the right places this strip all around the edge of the pouch. In both cases, the wallet is complemented by a handle, perhaps loosely inspired by a narrowed leather belt, which Nicolays assigned to the find (1882: Pl. IX.5).

Interpretations proposed by Birgit Heyerdahl-Larsen.
Heyerdahl-Larsen 1981: 37.

This interpretation is adopted by reenactors around the world, who use the wallet as almost the only surviving specimen of its kind found in Viking Age Norway. Sometimes both pieces are created to match the shape and material of the proposed reconstruction, other times the versions deviate more or less. The most noticeable differences between the individual trials are the dimensions, the contrast material and the material of the mouth reinforcement, which sometimes is wood or antler.

Version of the wallet made according to the current interpretation.
Maker: Dominik Schörkl (left), Královo řemeslo (right).

Version of the wallet made according to the current interpretation.
Maker: Bjorn This Way (left), norther73 (right).


However, on closer inspection, the current interpretation is burdened by several inconsistencies that concern three areas – shape, size and construction.


In order to be equipped with a reinforcing cuff at the mouth, Heyerdahl-Larsen suggests a shape that could be called a kidney-shaped. This is in contrast to Nicolaysen’s original drawing, which shows a more rounded shape and the absence of a cuff. If we take into account the degradation of the object at the time of discovery, which could simply have caused the distortion of the drawing, it can be assumed that the object was symmetrical and had the shape of an oval or an hourglass.

It is definitely worth mentioning that the openwork also looks somewhat differently and more fragile in Nicolaysen’s drawing.

Original Nicolaysen’s drawing. Nicolaysen 1882: Pl. IX.3.


The most common question of reenactors regarding the Gokstad wallet concerns dimensions. Here we cannot forgive the criticism of Heyerdahl-Larsen, who did not personally examine the wallet and considered it destroyed. If she made an effort to find a wallet, she would find not only that it existed, but also that its size deviated somewhat from the provided information. Let us remind that Heyerdahl-Larsen states the size of 13.5 × 8.5 cm.

To this day, about half of the front part of the wallet with one cut spiral has been preserved in a recognizable condition. Its photograph is digitized and stored in the central catalog Unimus under catalog number C10460 (Unimus 2020). From the attached scale, the accuracy of which was confirmed in a personal discussion by the author of the photograph Vegard Vike, it is clear that the size of 13.5 × 8.5 cm applies only to this one half of the object. Theoretically, the whole wallet would take on a size of about 17 × 13.5 cm, which in combination with a reinforcing cuff and loop is not very likely.

Nicolaysen gives a certain indication of how large the original object could have been in his original drawing, which has a 1:2 scale. We have checked other small objects from the book that are preserved to this day, with the result that the objects are drawn correctly or with only minimal deviations. Therefore, if we trust the scale, the front part of the bag measured approximately 15 × 11 cm according to the drawing. Unfortunately, it is not possible to verify the dimensions in the text part of Nicolaysen’s work, which does not contain this detail due to its very fast publication.

We believe that the difference between 17 × 13.5 cm and 15 × 11 cm is due to the fact that the photograph distorts the current state. The object is stretched and the area of ​​decoration is significantly stretched, which makes the object relatively long and wide. If it were possible to reduce the spacing between the openwork decoration to a minimum, as is the case with Nicolaysen’s drawing, we believe that it would affect the size. For this reason, the size of 17 × 13.5 cm cannot be taken too decisively and it is possible to lean towards a smaller size, approximately 15 × 11 cm. Although this dimension does not appear to be dramatically different from the original proposed size, it is a dimension that precludes some means of use. In this context, it is necessary to recall the remark of the conservator Vegard Vike that the whole object most likely underwent significant drying after the discovery, and thus shrinkage. The estimated size of the drawn state must therefore be taken with some reserve and can be considered as minimal. Another fact that must necessarily be considered is the turning of the finished product, which is certainly the case of the wallet from Gokstad.

Current state of the wallet. Unimus 2020.


We finally see the discrepancies in the description of how the individual parts of the wallet are constructed. Nicolaysen’s original text contains the following simple sentence, which only indicates that the leather was lined with colored fabric (Nicolaysen 1882: 47m):

m. divers fragments of leather with fine puctures […] doubtless parts of a purse which had been lined inside with coloured cloth, that has shown itself between the openings cut in the leather […].“

The current version of the Unimus catalog is the most infomative (Unimus 2020). In terms of language it corresponds to the period about 100 years ago. It is clear that Heyerdahl-Larsen based her infomation on this text:

‚Several piece of thin leather. Two of these pieces are oval, cut straight at one edge, 13.5 wide and 8.5 cm high, which correspond to each other and have been sewn together at the edges. Doubtless parts of a purse. Both pieces are made of two-layered leather, one of which has fine openwork patterning and was lined with textile. There was preasumbly yet another patterned piece. There were also straps, partially equipped with seam holes in the middle part.“

This report is absolutely essential for understanding the whole object, as it is the oldest description from when the object was more complete than it is today. We read about two two-layered oval pieces measuring 13.5 × 8.5 cm that are sewn together. Only one oval two-layered piece was decorated. At the same time, we learn that the piece probably had the same decorated counterpart. The text relates only to half of the wallet, which confirms the photograph of the object – the piece with the openwork is two-layered (the second, uncut two-layer part is not preserved) and corresponds in size.

Heyerdahl-Larsen, on the other hand, equates Nicolaysen’s drawing with the description of two two-layered pieces. Heyerdahl-Larsen interprets the two layer information as “two pockets” and transforms the information about the openwork counterpart into the idea that it was “probably decorated on both sides” (Heyerdahl-Larsen 1981: 37).

The belt was assigned to the wallet by Nicolaysen and, as noted above, the oval shape does not indicate the proposed use of the belts. Objectively speaking, it is not possible to confirm or refute whether the straps were in fact parts of the wallet.

New interpretation

In June 2020, we were contacted by the Danish craftsman Thomas Nichols from the Nichols Naturligvis workshop, who pointed out some of the above-mentioned discrepancies and expressed doubts whether the wallet was reconstructed correctly. We discussed the wallet in detail together and here we would like to publish our reading of the object, which – in our opinion – has been misunderstood to this day.

The shape of the wallet can be based on Nicolaysen’s drawing, which shows an approximately oval shape. The cross-section of the wallet must be based on a report from the Unimus catalog – the front had openwork, was two-layered and sewn to the back, undecorated two-layered part, which had to be a pocket. The back decorated part of the wallet was missing. Nicolaysen’s text and catalog show that the textile layer of contrasting fabric was placed between the two two-layered parts, where it prevented the deposited objects from falling out. It means that the wallet was originally folded and Nicolaysen’s drawing shows the wallet in an unfolded state. This reading is supported by 4 symmetrically placed holes, which in the previous interpretation were considered as decoration. The holes can be understood as a construction feature used to fasten a strap, which was apparently stretched in them. The same case may be the central elongated opening, which may not be a decoration but may have been used to store items in the wallet.

The closest analogy to our proposed interpretation is the wallet from Sigtuna (Sigtuna Museum 2019a; Sigtuna Museum 2019b). The object, found in Trädgårdsmästaren and dated to 1030-1050, consists of a front part measuring 14 × 11 cm, of a hourglass shape, with a simple geometric decor and a possible central opening. The inner parts of the wallet are missing, as with the current state of the Gokstad find.

The purse from Sigtuna. Sigtuna Museum 2019aSigtuna Museum 2019b.

Experimental reconstruction of the purse from Sigtuna. Author: Oleksii Malev.

The interior of the wallet offers several possible interpretations. In the visualization attached below, we propose three methods:

  1. both sides of the wallet are sewn together and the wallet is not openable. The interior is halved by one two-layered leather plate that creates two pockets. The textile layer prevents objects from falling out. The objects are inserted into the wallet through an opening on the upper edge.
  2. both sides of the wallet are left unstitched and the wallet is openable. The textile layer is covered with a leather part, which is then covered with two pockets. The opening on the upper edge is covered with textile.
  3. both sides of the wallet are left unstitched and the wallet is openable. The textile layer is partially covered with a leather part, which is then covered with two pockets. The opening on the upper edge is covered with textile. The space between the leather part and the textile can be used to store items, so the wallet has a total of four pockets.

Unfortunately, we do not know the nature of the textile used. The only information we have is the color of this textile layer, which created a contrast. It is not impossible that it may have been the silk, which is represented by the other find from Gokstad mound (Vedeler 2014: 41-2).

The position of the belt, which was assigned to the find by Nicolaysen and subsequently interpreted by Heyerdahl-Larsen, cannot be estimated. We should not rule out the possibility that these fragments do not belong to the wallet at all.

Visualization of the new interpretation

In collaboration with the reenactor and saddler Thomas Nichols from the Nichols Naturligvis workshop, we have prepared a visualization of the described interpretation, which proposes the construction step by step. The author notes that the production of the wallet takes about 2.5 hours, which is a significant acceleration compared to the 9-11 hours that took the production of the wallet of the current interpretation.

Pictures of the variants can be easily downloaded via the following link:

Variant 1


Variant 2


Variant 3


Revision of the find would not be possible without observant eyes and a determined mind of Thomas Nichols from Nichols Naturligvis workshop, who created functional models of the newly interpreted wallet and whom I warmly thank.

Here we will finish this article. Thank you for your time and we look forward to any feedback. If you want to learn more and support my work, please, fund my project on Patreon or Paypal.


Heyerdahl-Larsen, Birgit (1981). Gokstadhøvdingens pung = The Gokstad chiefstain´s pouch. In: Wexelsen, Einar (ed). Gokstadfunnet : et 100-års minne = The Gokstad excavations : centenary of a Norwegian Viking find, Sandefjord, 36-7, 104-5.

Nicolaysen, Nicolay (1882). Langskibet fra Gokstad ved Sandefjord = The Viking-ship discovered at Gokstad in Norway, Kristiania.

Sigtuna Museum (2019a). Veckans föremål. In: Sigtuna Museum & Art. Visited 19th March 2020, available from:

Sigtuna Museum (2019b). Uppdatering av Veckans föremål. In: Sigtuna Museum & Art. Visited 14th June 2020, available from:

Unimus (2020). C10460. In: Unimus [online]. [2020-06-14]. Available from:

Vedeler, Marianne (2014). Silk for the Vikings, Oxford – Philadelphia.

The forms of Norwegian sword grips

While self-learning about swords from Early Medieval Norway, I noticed quite a number of specimens having well-preserved organic remnants of hilts and sheaths. Because normally these components do not remain, I considered useful to gather these materials into coherent articles to enrich sword enthusiasts. In this article we will follow up on Norwegian sword hilts from the Viking age, however we think that the Norwegian material can be used for pointing out all the sword hilt types used in Early Medieval Europe.

We start by quick recapitulation of how we define the hilt and handle of these swords. The sword is composed of a blade and hilt components which are mounted on the blade’s tang. The hilt components are the lower guard (fremra hjaltit) and a pommel, which can be one-pieced or two-pieced. In case of the one-pieced pommel, the tang goes through and is riveted at its top. However, in case of the multi-pieced pommel, the tang is usually riveted to the upper guard (efra hjaltit) and then the cap of the pommel is riveted to the upper guard by two rivets. Uncovered part of the tang between the lower guard and the pommel provides the space for a handle (meðalkafli). Now we will explore the ways of covering the tang to enable comfortable and smart use of the sword.

Viking Age sword terminology. Created by Jan Zbránek and Tomáš Vlasatý.

During a fairly thorough exploration of the Unimus catalogue I was able to find four main hilt forms. We can notice that most of the hilts are anatomically shaped, widening towards the lower guard.

Wooden handle
A tang covered with a wooden handle of an oval cross-section – that seems to be the most common variant used in the Viking Age, which also has great variability. The handle could be made of two identical scales, a cylinder with a burnt-in gap or a cylinder with a cutting that would be covered with a narrow piece of wood when mounted on the tang. The material seems to be the wood of broad-leaved trees (i.e. T16054 and T20736, and the pre-Viking B4590 seems to have a birch handle as well). In contrary to Pre-viking periods, the profiling of the handles in terms of finger copying bumps is not documented. Some swords seem to have only an unwrapped wooden handle, in other cases the wooden handles are wrapped in leather, fabric, metal or a combination of these.

  • Leather wrapping
    We have evidence that the wooden base was wrapped with a leather cord (C57001) or a strip of leather (T14613). The shape cannot always be reconstructed exactly, at least in one case the wrapping leather does not have a specified shape (C23127, Ts2954). Identical leather wrapping can be found on Swedish and Icelandic swords.

  • Cloth wrapping
    We can sometimes detect thread (S3821), textile strap (B5161) or cloth plus iron wire wrapping (T3107). Some of the finds are wrapped in unspecified textile (S11782, T12962, T21998). Identical methods of cloth wrapping can be found on Swedish swords as well.

  • Wire wrapping
    Silver, gold or copper alloy wire was a quite popular and very spectacular option for wrapping (C5402, C22138, C23486, C58882, T19225) as manifested on S and Æ types. This variant is also mentioned in written sources (vaf), specifically in the context of elites around the ruler and rich farmers (Falk 1914: 23). As we can see, the wrapping was typically executed with orderly separation of thin wires and two pairs of coiled wires opposite to each other, thus creating the fishbone effect. The wires are often entwined into curls of thicker wire at the ends of the handle. Wooden handles were quite minute under the wire, making the resulting handle rather subtle. This can arise some questions regarding possible special designation of such swords, for example combat swords fit for stabbing (personal debate with Roland Warzecha).

  • Metal ferrules on the handle
    Usage of bronze pre-cast or plated ferrules at the ends of wooden handles was equally popular (B1481, B11477, C1194, C1977, C5464, C8095, C9981, C11301, C16107, C18494, S5371, T8257, T16054, T20913). Pre-cast ferrules are crown-shaped and their tongue-like protrusions often depict animal or humanoid heads. These ferrules are probably mentioned even in written sources under the name véttrim (Androshchuk 2014: 31). Some hilts have simple ferrules spread on the inner surface of the handle (B878, B11477).

rukojetiDiverse variants of Norwegian sword handles.
B8118, C57001, T3107, C58882, T16054.

Swedish analogies. Taken from Androshchuk 2014: 104-105.


Antler handle
As far as I know, there was only one sword in the Viking age (S2453) with its grip made of antler scales. Handles made of this material are very rare in neighbouring Sweden too, where only two specimens (Androshchuk 2014: Jä 12; Holm 2015) were found. The antler scales of Swedish swords were riveted on the side with tiny iron rivets.

Antler handles from Norway and Sweden.
S2453 (left), SHM 12426 (right).

Straw / bast wrapping
According to the Unimus catalogue, a single-bladed sword was found at Tussøy (Ts3639) whose handle was wrapped in straw or bast. This modification seems to be completely unique and I know of no parallels to it. Due to insufficient description, we can provide no detailed information. In addition, Sveinulf Hegstad, the photo archivist University of Tromsø, provided me with a current photo of the object and there is no organic trace left.

Metal handle
Pre-cast or forged handles are found on some of Petersen’s type D swords in Norway (i.e. B5774, C4072. C8095, C24887, T14309). These swords, being among the heaviest of all Viking swords, can be dated to 800-950 AD. They are composed of triple-lobed, two-pieced pommel, guard and typically also a metal handle. These handles are massive products of metal casting or smithing and their surface is covered with geometric or animal decor. The lower layers of the profiled decoration are decorated with copper alloy, the upper layers with silver. These handles are also sometimes decorated with ferrules on the handle tops (véttrim). According to Petersen, there were 11 swords of this type in Norway in 1919 (Petersen 1919: 70-75), while Hernæs filed up to 16 specimens by 1985 (Hernæs 1985). The Petersen type K sword from Nodland (S4262) also uses copper alloy grip and probably copies the type D in this feature.

rukojeti-kovoveD type swords with metal handles.
B5774, C4072. C8095, C24887.

I hope you liked reading this article. If you have any question or remark, please contact me or leave a comment below. If you want to learn more and support my work, please, fund my project on Patreon or Paypal.


Androshchuk, Fedir (2014). Viking Swords : Swords and Social aspects of Weaponry in Viking Age Societies, Stockholm.

Falk, Hjalmar (1914). Altnordische Waffenkunde. NVAOS. No.6., Kristiania.

Hernæs, Per (1985). De østnorske sverdfunn fra yngre jernalder : en geografisk analyse. Magistergradsavhandling i nordisk arkeologi – Universitetet i Oslo, Oslo.

Holm, Olof (2015). A Viking Period sword from Skäckerfjällen with a decorated antler grip. In: Fornvännen 110:4, 289-290.

Petersen, Jan (1919). De Norske Vikingesverd: En Typologisk-Kronologisk Studie Over Vikingetidens Vaaben, Kristiania.

Gokstad belt recreations


Then and now : the mound after the opening and the current state.

Dear reader, welcome back on this site that is dedicated to research and reenactment!

This time, we will examine belt components from Gokstad mound, Southern Norway. Being covered with 50×43 meters big mound and consisting of a richly furnished ship, the grave is one of the most well-known Scandinavian burials (more here and here). The buried person was probably a man of high rank that was connected to ruling family. Thanks to dendrochronological analysis, it was found that the timber for the burial chamber was cut in the first decade of the 10th century, and therefore the whole grave can be dated to this period (Bonde – Christensen 1993).

Even though the grave was robbed and all weapons and valuables were presumably taken, the presence of organic remnants – like skeletons, leather and wooden objects – as well as some cast products, makes the grave significant. However, the only scientific overview of the find was published by Nicolay Nicolaysen in 1882. It might seem some objects are not even treated in the book, while others are not depicted or described, but we have to realize that the mound was re-opened several times, namely in 1925 and 1928/9. From around 1950 onwards, Gokstad grave has been given academic attention several times, that covered bone, wood and metal analysis, detailed scanning of wooden objects and non-destructive documentation of the mound and near landscape. This delicate work has brought some light into how colourful the grave was originally (for example Bill 2013).


The grave of Gokstad recreated. Made by Ragnar L. Børsheim,

Among the finds, there were also many belts components. Before the experimental part of this article, it has to be said that it is not able to determine the sets, nor which components could be waist-worn and which were used as parts of horse bridles. That makes reconstruction extremely difficult, virtually impossible. To sum up, there are at least six belt buckles, at least nine strap-ends, at least seventy-four mounts of eleven different kinds and at least three belt slides. The complete list can be seen or downloaded here. Given the fact the burial consisted of twelve horses, eight dogs, several birds etc., it is very probably the most of belts belonged to animals. In the text below, you can read two different approaches of experienced reenactors and owners of custom-made Gokstad belt recreations. They both try to portray Norwegian high rank men from the 9th/10th century.


Reconstruction of the bridle from Borre. Taken from


Reconstruction of the bridle from Gokstad.



Selection of belt components from Gokstad. Taken from Nicolaysen 1882.

joschJosch Weinbacher

Mannschaft der Ormrinn Brands, Austria

Belts are a crucial parts of reenactor kits. I consider them to belong to the basics, that everyone should get for a start, next to a tunic, trousers, shoes and a simple everyday-use knife. For a lower class character basically everything that can bind the tunic at the waist can serve as a belt. There is, hovever, a tendency towards richly decorated belts, and reenactors often purchase beautifully looking belts with rich fittings, even before doing proper research. I was no different in the beginning, I have to admit. When I started, I bought the first „viking-style“ belt, labelled so because of an overall nordic style, but absolutely not fitting to the region and time I wanted to depict (Norway in the 9th century). It was, in fact, not nordic, nor even early medieval at all, as I found out later.

I could have avoided that by doing my research, but also by taking smaller steps first. A simple D-shaped buckle would have served me perfectly, as I now recognize, and in my opinion even a simple leather strap, a piece of hemp rope or a pleated band would have been sufficient.

After a while, when my ambitions grew and my methods of research got better, I recognized that the issue with belts was a big one, because of a simple fact: tunics, trousers, shoes and knifes are somewhat generic in their overall look, it is hard to specify a reenactors region and timeframe by them alone. The fittings of a belt, however, can identify a person, if they are shaped according to a specific find. That is not only true for belts, but for jewellery in general. That’s way you can easily spot for example a brooch from Gotland on Norwegian woman’s apron, and it can be supposed she did not do her research properly. For belts it is much the same, regions and timeframes get mixed and mingled with others or are chosen wrongly, horsegear appears on people, and even unintended crossdressing can happen. Therefore, I decided that I had to purchase something that would fit the region and timeframe our group depicted better. The Gokstad ship-burial seemed obvious in that regard, because I am the leader of our group and was supposed to show some wealth in my kit.

This was actually of a great difficulty for me. Showing wealth in your kit is, to some extent, forcing you to be wealthy in reality too. Of course a modern recreation of a period piece does not match the worth of the original, but they can be quite expensive anyways. Needless to tell any reenactor that this hobby is an expensive one, I am sure.

When I decided to get myself Gokstad belt, I checked out some artisans who cast belt-fittings, located in Germany. The prices were stunning, and in the end I went along with a kind of poor recreation from an e-shop, that only featured the buckle and strap-end I desired, but no further ornaments, and it was smaller in size than the original. I went along with that for some years, but I was never fully satisfied. It was by mere chance that I later discovered a maker in Poland, who had quite reasonable prices and sold belts with Gokstad fittings. The assambling of the belt was not perfect, because the fittings were placed in a way, that they would be visible if one used the famous belt-knot that is widely accepted in reenactment, but for which there is not real evidence I have knowledge of. So I ordered the fittings only, and intended to assemble the belt myself.

Meanwhile I asked one of our group members, who had allready gained some experience in dying leather with period ingredients, if he could dye a strap for my belt in a bright red, making the finished piece more imposing. He came up with a recipe he found in the Mappae Clavicula, speaking of red wine and kermes. Cochineal was used as a replacement for kermes, again a matter of finances. The result was great. The belt did not become bright red, as intended, but took on dark, almost purple red, much like the colour of wine. For me, it is mostly that colour that makes the belt so great. When the ormaments arrived in the end, I only had to assamble the whole thing. Now I’m finally satisfied with my attire, even if the belt is not yet finished, since I’m still lacking one specific fitting, that I will add when I manage to find it. So my journey to a beautiful belt was a long one, and I have not yet fully completed it, but I am happy that my kit is again a bit improved. And that is, by all means, a process, that can never really end.


tomasTomáš Vlasatý

Marobud, Czech Republic

During my reenactment “career”, I have had about five or six belts. Some of them were done with pure fantasy, others were based on particular finds. In the beginning of 2016, I started to feel the need for a new belt, that would fit to my 10th century Norwegian impression. To be honest, it is not so easy to find a well-preserved belt, consisting of a buckle and a strap-end, in the region. Therefore, I decided for Gokstad.

My incredibly skilled friend Jan Bana from Storrvara took the task and made the set to order. During the process, he kept me updated by photos, so I could make some correction online. After several months, the bronze set was done, for a really reasonable price. The set consists of a buckle (C10437), a strap-end (C24239c) and twelve mounts (6×C10445 and 6×C10446). My friend and fellow Jakub Zbránek mounted the components to an impregnated belt for me.

It is true that my choice was quite hasty and motivated by the urge of recreation of unique objects. Indeed, some components are, to my best knowledge, the first imitations after 1100 years. Due to my decision, we were forced to make the buckle a bit smaller than the original, with a bronze tongue and without a folded sheet; the find from Hedrum (T1620) can be an analogy, when it comes to reconstruction. Another mistake is that no component is gilded. The biggest fault, however, is the usage of mounts, that were, with high probability, parts of horse bridles. If I spent more time doing the research, I would save money, and more importantly, my kit would be more accurate. On the other hand, my mistakes encouraged me to write this article. The fact that I was wrong is very important for me and my future progress. I am sure that I am going to order a new one in some time, a belt that would be more accurate and that could be called “a replica”.


Before the very end, let me express my thanks to Josch Weinbacher. In case you found this article inspiring, feel free to share it in your community or let us know. For any questions or notes, please, use the comment board below. Love the past, enjoy the present and look forward to the future! If you want to learn more and support my work, please, fund my project on Patreon or Paypal.

  • Bill, Jan (2013). Revisiting Gokstad. Interdisciplinary investigations of a find complex investigated in the 19th century: In: Sebastian Brather – Dirk Krausse (ed.), Fundmassen. Innovative Strategien zur Auswertung frühmittelalterlicher Quellenbestände, Darmstadt: Konrad Theiss Verlag, s. 75–86.
  • Bonde, Niels – Christensen, Arne Emil (1993). Dendrochronological dating of the Viking Age ship burials at Oseberg, Gokstad and Tune, Norway. In: Antiquity. A quarterly review of archaeology vol. 67, 256, p. 575–583.
  • Nicolaysen, Nicolay (1882). Langskibet fra Gokstad ved Sandefjord = The Viking-ship discovered at Gokstad in Norway, Kristiania.