Introduction
In this second instalment of the paper dedicated to Petersen Type T swords, our focus shifts towards the second sub-group of the type under discussion. In the first part, we analysed swords of the first subtype, accompanied by a brief commentary on groups of swords whose morphological and decorative characteristics closely resemble this subtype. Within these, we distinguished not only atypical sword forms classified under type T1, but also diverse variants originating within the Eastern Baltic region (Viskupič 2026a).
In the following section, we maintain the structure of the article, dividing it into distinct chapters. Our primary objective here is to map the second subtype as defined by Jan Petersen from the perspectives of construction, decoration, distribution, and dating. In keeping with standard practice, the conclusion is supplemented by a catalogue of swords presented in tabular form.
Description of Structural Components
Before proceeding to the analysis of the individual components that comprise a complete sword of the second Petersen Type T subtype, let us briefly recapitulate the fundamental knowledge regarding the first subtype. Jan Petersen recorded six Type T swords from the territory of Norway, which he divided into two variants based on their style of decoration and the shape of the hilt’s structural components (Petersen 1919: 150–153). The first variant was defined as having massive guards and a pommel covered in silver. Within this subtype, he identified two methods of decoration: the first, less common, zoomorphic style consisting of intertwined animal bodies; the second variant features so-called pitted decoration, characterised by circular cells whose bases are filled with twisted silver wire threaded through channels beneath the iron surface of the hilt components. The more widespread subtype T1 with pitted decoration is considered an evolutionary form of the chronologically earlier Type E swords – specifically the E-2 subtype – with which it shares an identical pitted decoration and twisted wire inside the cells (Kainov – Novikov 2024: 199). Both variants of the first subtype also feature a method of mounting the pommel onto the blade tang identical to that of Type E.
Generally speaking, this type does not belong among the numerous sword variants of the Viking Age, and its geographical distribution is restricted to a relatively small number of sites, with the most significant concentration occurring primarily in Norway and the Old Rus’ environment (present-day Russia and Ukraine). Two isolated finds have been recorded in Finland and Hungary. On Gotland, an unfinished upper guard was discovered, which Fedir Androshchuk identifies as the upper guard of a Type T1 sword (Androshchuk 2014: Go 138*).
Type T1 swords with pitted decoration most likely served as the inspiration for the emergence of a diverse group of swords in the Eastern Baltic region, which is characterised by the same assembly method of a two-part pommel with a three-lobed pommel cap and pitted decoration. Although these swords are assigned to Type T, or in some cases considered direct forms of Petersen’s first subtype, they display certain differences in the shape of their pommels and crossguards, as well as in their decoration – such as polychrome decoration or the absence of wire at the base of the pits. Consequently, they must be viewed as local forms of swords with unique ornamentation.
The first and second variants of Petersen’s Type T represent two distinct forms of swords with differing evolutionary lines; however, they share, most notably, the shape of the pommel, which is constructed from two separate components: the upper guard and a three-lobed pommel cap. Nonetheless, on three swords of the first variant, we encounter a pommel cap divided into five segments. In addition, a common element shared by both subtypes is the presence of inlaid decoration. These morphological and decorative features thus link subtypes T1 and T2 with other sword types, namely Types R, S, V, and Z (Hošek – Košta – Žákovský 2021: 319; Martens – Astrup 2021: 26). Jan Petersen considered Types R, S, and T with blades decorated with the Ulfberht inscription to be swords originating from Frankish workshops; at the same time, he added that the origin of Type T as a whole remains uncertain (Petersen 1919: 153).
As stated in the first part, neither the first nor the second subtype is included in Geibig’s combinational typology, given that not a single find of these swords is known from the territory of former West Germany. The second variant, which is discussed in detail below, is likewise absent from the classification by Oluf Rygh. It may further be added that within Jakobsson’s typology, the researcher assigned Type T swords to the “second principle”, which represents a group of swords featuring narrow, three-lobed, and plastically raised pommel caps (Jakobsson 1992: 36–37).
Fig. 1. Swords with a three- and five-part division of the pommel cap:
Type R, unknown locality in Innlandet county, Norway – C257 (adapted from unimus.no);
Type S-T2, swords from Dneprostroi, Ukraine (after Kirpichnikov 1966a: Fig. 4);
Type V, Shestovytsia, Ukraine (after Kulakov 2019: Fig. 3.3);
Type Z, Bengtsarvet/Häradsarvet, Dalarna, Sweden – SHM 22293 (adapted from historiska.se).
Type T2 swords are characterised by lighter, narrower crossguards and a three-lobed pommel cap. Within this type, two methods of polychrome decoration featuring geometric patterns can be distinguished (Petersen 1919: 151; Androshchuk 2014: 76). The study presented below is therefore based on the criteria defined by Petersen for Type T2 swords, incorporating assessments from more recent studies while taking into account the morphological attributes of the hilt’s structural components and their decoration (Kainov 2012; Androshchuk 2014; Maczek et al. 2024).
Furthermore, this analysis also examines swords that display some of these defined characteristics but cannot be conclusively assigned to the type under discussion due to their poor state of preservation and/or missing decoration; these specimens have been designated as ‘uncertain Type T2’.
The Pommel
The pommel consists of two hollow iron components: the upper guard (base) and a three-lobed pommel cap. In all known examples, the upper guard appears straight in frontal view, often featuring side edges that slope sharply upwards towards the pommel cap. Specimens with rounded terminal ends have also been recorded. In plan view, the upper guards possess a lenticular shape, being thickest in the centre and tapering towards the ends, with rounded edges. The centre of the upper guard is perforated by a rectangular slot designed to accommodate the blade tang. On both the left and right sides, closer to the lateral edges, there are circular holes intended for the rivets that secure the upper guard to the pommel cap.
The structural components of the pommel (as well as the lower crossguard) exhibit morphological variations, a feature also noted by Alexandra Shchedrina, who states that some specimens may feature more massive components while others are thinner (Shchedrina 2023: 94).
Regarding the continuity of form in the pommel caps across individual variants, their differing characteristics are noteworthy. In general, we encounter steeply or gently concave lateral lobes stylised into animal heads, as well as high, triangularly shaped lateral segments with a rounded apex. Many of these lateral segments are characterised by prominent ridges. In some specimens, it is noticeable that the central lobe is more semi-circular, closely resembling Type V, though featuring more massive proportions. Conversely, in other cases, it is not particularly high but rather low, which some researchers suggest may indicate transitional variants between Type V and Type T2 (Maczek et al. 2024: 293–294). The division of the pommel cap between the lobes is achieved by obliquely oriented, deeply grooved vertical channels, which were typically filled with twisted silver wire arranged in three parallel rows. Based on our observations, only one sword survives with this decoration intact, originating from Utgården in the Telemark region of Norway (C1648).
A similar arrangement was used for the junction between the upper guard and the pommel cap, a feature observable on most Viking Age swords with a two-part pommel. The twisted wire could generally be arranged in three superimposed rows, as demonstrated by finds of Type V swords from Kyiv and Birka (Androshchuk – Zotsenko 2012: Fig. 61; Androshchuk 2014: Pl. 142a), or in two rows, as documented on a Type T1 sword from Dneprostroi (Komar 2014: Fig. 8.1). Wrapping with a single twisted wire was also not uncommon, as can be seen on a Type V specimen from Østabø in the Rogaland region of Norway (S2266/a), although it is more frequently encountered on Type H/I swords (Kainov – Novikov 2024: Figs. 214.4, 219.1, 2, 4, 221.1–2). Within the type under study, however, no specimen has been recorded that retains the wire separating the pommel cap from the upper guard. Furthermore, the absence of oblique grooves between the lobes is documented in three cases. On one of these, the three-part division of the pommel cap and the boundary between the cap and the upper guard are rendered using inlaid twisted copper and silver wire forming a zig-zag pattern, alongside copper and silver wire arranged in a herringbone pattern (see the Decoration chapter below). A similar decorative style can also be found on certain Type V swords, such as those from Gnezdovo (destroyed barrow) and Pidhirtsi (Kainov 2012: 46, 50, Fig. 36; Viskupič 2025).
Fig. 2. Schematic diagram of several examples illustrating variations in the pommels of Petersen Type T2 swords. Upper row from left to right: Utgården, Kjølstad Vestre, Norway; Košice – Krásna nad Hornádom, Slovakia; lower row from left to right: Novinka and Vakhrushevo, Russia; visualisation of the upper/inner and lower section of the upper guard (drawing by author).
The method of mounting the pommel differs significantly in the T2 subtype compared to the first variant. While subtype T1 swords have their pommels secured as a single unit by the blade tang (and two small pins at the lateral edges of the pommel cap and upper guard), which corresponds to Geibig’s structural Variant I, the fastening of the T2 subtype pommel is achieved by a curved rivet resembling an inverted U-shape (∩). In Geibig’s structural typology, this corresponds to Variant II (Geibig 1991: Fig. 24). This method of assembly can be identified on three swords based on X-ray radiographs (Maczek et al. 2024: Fig. 4; Janionis 2025: Fig. 46; unimus.no – C1648). The entire construction relies on the blade tang being inserted into the interior of the pommel cap, through a small rectangular slot at its base, where its end is riveted. On the sword from the Slovak site of Krásna nad Hornádom, the X-ray radiograph clearly reveals that the end of the tang is fixed here by a cylindrical metal washer (Môc 2018: 223; Maczek et al. 2024: Fig. 4).
The curved rivet was inserted into the interior of the pommel cap, likely using solder, which was identified as brass in the case of a Type V sword from Olomouc (Frait 2006: 98). The rivet passed from one side to the other, secured inside the pommel cap at the apex of the central segment, following its contours. After the pommel cap was fitted onto the upper guard, the ends of the rivet were inserted into holes in the upper guard, where they were subsequently peened (Môc 2018: 223) and covered by two circular iron lids, a feature observable on two swords from Russia (Kainov 2012: 46, Fig. 31; Shchedrina 2023: Fig. 3.2). The rivet in the pommel of the sword from Krásna nad Hornádom is estimated to have a diameter of approximately 3 mm (Môc 2018: 223). This constructional method corresponds to Geibig’s second variant (Geibig 1991: 99, Fig. 24). Apart from Type T2, this feature is encountered primarily on the closely related Type V (Androshchuk 2014: 78–79; Petri 2017: 139, Fig. 1; Hošek – Košta – Žákovský 2021: Fig. 107b–c), but also on Types H/I (Androshchuk 2014: 57; Kainov – Novikov 2024: Figs. 68.1, 3–4, 210.3) and Type N (Androshchuk 2014: 69).
Fig. 3. Close-up of X-ray radiographs of Type T2 (a–c) and Type V (d) sword pommels:
a. Viešvilė, Lithuania (after Janionis 2025: Fig. 46);
b. Utgården, Seljord, Telemark, Norway (adapted from the unimus.no portal);
c. Košice – Krásna nad Hornádom (after Maczek et al. 2024: Fig. 4);
d. Wiskiauten, grave 11 (after Petri 2017: Fig. 5).
Fig. 4. View of the lower part of the upper guard showing the placement of rivets covered by a circular lid and a protruding rivet on Type T2 swords from Gnezdovo and Novinka, Russia (after Kainov 2012: 46, Fig. 31; Shchedrina 2023: Fig. 3.2).
A noteworthy structural feature of the pommel can be observed on the sword from Momrak, Fyresdal, Telemark county in southern Norway (C28352). According to the online portal of the university museums of Norway, this sword is closely related to the specimen from Kosgaarden in Innlandet county, eastern Norway (C3210), which Jan Petersen assigned to the first variant (Petersen 1919: 150, Fig. 119). According to the portal’s records, the sword has been preserved without decoration; however, distinct silver residues in the form of granules are visible in some areas, suggesting that the object was likely exposed to fire in the past. A research team consisting of Maczek, Môc, Dvorák, and Beronská also inclines towards its classification under subtype T1, arguing that the pitted decoration was lost due to the weapon’s exposure to fire (Maczek et al. 2024: 293). Both the crossguard and the pommel exhibit distinct vertical incisions on their surfaces, which is rather characteristic of swords of Petersen’s second variant of Type T (for more on this topic, see the Decoration section). Also typical of Type T2 is the appearance of the pommel and crossguard with reversely sloped edges. The lateral segments are characterised by prominent ridges and a steeply concave section. On its underside, between the rectangular slot for the tang and closer to the terminal ends, the upper guard features two larger circular holes, which could have served to thread the rivet connecting the pommel cap to the upper guard. Their placement corresponds to other holes or rivets detected on Type T2 swords (Kainov 2012: 46, Fig. 31; Shchedrina 2023: Fig. 3.2; Maczek et al. 2024: Fig. 4; Janionis 2025: Fig. 46).
Another distinct element indicating that the Momrak sword belongs to the second subtype is the thinner layer of iron forming the upper edge of the upper guard and the lower edge of the pommel cap. In some T1 specimens, a pronounced thickening of this part of both components is clearly visible, particularly near the lateral edges where the circular pits for the pins connecting the upper guard and pommel cap may be located (Pienemäki 2016: Appendix 8.4[22]; Shchedrina 2018: Fig. 1; Kainov – Novikov 2024: Fig. 189.1–2). Despite this, the X-ray radiograph clearly demonstrates that the pommel cap and the upper guard are connected by the blade tang. Photographs from the Unimus portal also show a perforation in the upper section of the pommel cap through which the tang passes. In the internal part of the pommel cap, several photographs, including the X-ray radiograph on the Unimus portal (1, 2, 3 – edited), reveal a distinct preserved band that could represent the remains of solder (Fig. 3a–d).
Taking into account the morphological characteristics of the pommel and crossguard, it can be assumed that the sword belongs to Type T2 and that the perforation of the pommel may have been a subsequent modification following damage to the sword, or the pommel may have been damaged during the assembly of the individual components (personal communication with Tomáš Vlasatý). Similarly, the possibility that the pommel and crossguard originated from a different sword and were later mounted onto a newer blade cannot be ruled out.
Fig. 5. The sword pommel from Momrak:
a. detailed view of the hilt; b. X-ray radiograph of the sword; c. view of the internal and upper part of the pommel cap; d. view of the internal and lower part of the upper guard (a–d, adapted from unimus.no – C3210).
The length of the pommel caps ranges from 6.6 to 8 cm, their height varies between 3.1 and 3.88 cm, and their thickness ranges from 1.7 to 3.5 cm. The upper guards range from 6.5 to 9.4 cm in length, between 1.4 and 2 cm in height, and from 2.2 to 3.91 cm in thickness (all data based on personal observation). Fully preserved pommels are recorded on 23 specimens. Two isolated finds of pommel caps are also known from Latvia.
The Blade Tang
The iron blade tang has survived in a complete or fragmentary state on approximately twenty-four swords, four of which retain their organic or metallic wrapping. Two swords from the Finnish site of Köyliö, Vanhakartano feature a tang covered by a wooden grip. In addition to the wood, a poorly preserved textile covering – a woollen twill fabric that was unfortunately destroyed during conservation – was detected on the specimen with inventory number KM 8602:130; another part of its grip consisted of a wrapping made from leather strips approximately 7 mm wide (Tomanterä 1972: 24; Moilanen 2015: 387; 2018: 124). The second find (KM 8602:141) was preserved with a fragment of a wooden grip on the lower part of the blade tang (Moilanen 2018: 125).
Fig. 6. Preserved grip coverings of Type T2 swords from the Finnish site of Köyliö, Vanhakartano:
a. KM 8602:130 (after Moilanen 2018: 124); b. KM 8602:141 (after Moilanen 2018: 125).
Apart from organic materials preserved on Viking Age sword hilts, wire wrapping is also documented. Within the studied group, this feature can be observed on only one sword, which originates from Krásna nad Hornádom. The wrapping is continuous, using a smooth wire along the entire length of the tang, and it can be noted that it is spliced twice (Môc 2018: 223). The wire is 0.6 mm thick and consists of a copper alloy, most likely brass (Môc 2018: 223; Maczek et al. 2024: 280). Swords with hilt wire wrapping made from yellow non-ferrous metal are documented on only a few specimens, such as a Type V sword from the Estonian site of Raatvere (Kiudsoo 2019: 50) or a sword from a private Croatian collection (Kovač 2003: 16). This method of wrapping corresponds to Type 2.1a according to Vlasatý’s classification and appears to be the most frequently used variant on swords from Northern, Central, and Eastern Europe (Vlasatý 2020b).
Fig. 7. Detailed view of the hilt wrapping of the sword from Košice – Krásna nad Hornádom (after the SNM archives).
In one case, a sword find from Latvia features a wooden grip covered by a narrow metallic sheet strip, probably made of a copper alloy, as evidenced by photographs of the object published on the swordmaster.org online forum (photograph of the wrapping). Similar grips wrapped in metal sheet occur very frequently in the Baltic environment and can be found on swords of various types: Doles Vampeniešu – Type L (Tomsons 2018: 54, Fig. 18); Aglonas Kristapiņi – Type V (Tomsons 2018: Fig. 32.2; Viskupič 2025: Fig. 5); Palanga – the antenna type according to Kazakevičius (Vlasatý 2020b; Janionis 2025: Cat. No. 78). In addition to swords, this type of grip covering in the Baltic region is also documented on the shafts of narrow, lightweight axes or in the case of several spearhead finds (Vlasatý 2022a).
The length of all completely preserved tangs ranges from 8.5 to 9.5 cm, which corresponds to the width of a human palm and indicates custom-made production (Androshchuk 2014: 105; Vlasatý 2020a).
The Crossguard
The crossguards of this type are morphologically similar to the upper guards, and their heights also more or less correspond. In frontal view, the crossguards have a straight shape; in plan view, they are lenticular, being thickest in the centre and tapering towards the ends to form rounded edges. A characteristic feature of some specimens is the oblique sloping of the side edges towards the blade. A similar trait can be found on swords of the first subtype, or it is encountered on various variants of local late Viking Age swords originating from Estonia, Finland, the Eastern Baltic, or the former Kyivan Rus’ (the silver-plated type, Baltic forms of swords designated as Type T, the Krasnianka type, etc.). The upper part of the crossguard features an oblong, rectangular slot in the centre through which the blade tang passes. Conversely, the lower side is equipped with a narrow opening that follows the lenticular shape of the component, through which the hilt tang was threaded and where the blade was secured.
A noteworthy example is the sword from Finland (KM 8602:141), which retained a woollen twill textile on its crossguard (Tomanterä 1972: 26, Fig. 5). The thickness of the weft thread was estimated at 0.5 mm and was spun in the clockwise direction (Z-spun) with a density of approximately 8 threads per cm (Tomanterä 1972: 26). The warp thread was spun in the counter-clockwise direction (S-spun) with a total thread thickness of approximately 0.9 mm, while the thickness of an individual thread was about 0.5 mm, with a weft density of 8 threads per cm (Tomanterä 1972: 26). The textile has not been preserved; it was destroyed during the conservation of the sword (Tomanterä 1972: 26).
The length of the shortest recorded crossguard is 9 cm, whereas the longest reaches 12.57 cm. The height varies between 1.5 and 2.53 cm, and the thickness ranges from 1.3 to 2.7 cm (personal observation). Preserved crossguards are documented in 25 finds.
Fig. 8. Schematic visualisation of crossguard form shapes of Type T2 swords: top left and right, examples illustrating variations in crossguard shape; bottom left – schematic representation of the upper side of the crossguard; bottom right – schematic representation of the underside of the crossguard (drawing by author).
The Blade
At the very outset, it is necessary to mention that due to the complexity of the subject of independent blades – whether from the perspective of their manufacturing process or the presence of inscriptions and marks – the analysis of Type T2 sword blades below will be restricted exclusively to a brief commentary. This will consist of metric data, an overview of all variants of inscriptions and/or marks, and the identification of organic materials from the scabbards in which the sword blades were sheathed.
The blade has been preserved in 24 cases. Of these, only three can be considered completely preserved with relatively minor damage. On another three swords, the blade is largely intact, albeit with minor damage in the area of the tip and along both cutting edges. One sword survives with a complete blade that is currently broken into four pieces, with the tip section being bent. In the remaining cases, the blades are preserved in a rather fragmentary state. We were able to verify the total length in only three instances, with the longest blade reaching 81 cm and the shortest 71.2 cm (Kochkurkina – Summanen 2021: 251; Maczek et al. 2024: 282). The average width of the blades below the crossguard ranges between 5 and 5.6 cm, though we also find a blade with a width exceeding 6 cm (personal observation; Tomsons 2018: Tab. 1, Cat. No. 47; Kochkurkina – Summanen 2021: 251). All Petersen Type T2 swords are double-edged. According to Artūrs Tomsons, a Type T2 sword with a single-edged blade allegedly originates from Latvia (Tomsons 2018: 64, Tab. 1, Cat. No. 46). Given that this sword survives only with a crossguard and an upper guard that clearly bear no decoration – or at least none that we could verify – it is assumed that the sword likely does not belong to the type under study.
On all monitored blades, with the exception of those found with the remains of the structural parts of a scabbard, a central fuller is discernible. In 13 cases, this fuller is provided with inscriptions and/or marks.
The most numerous group of inscriptions on Type T2 sword blades consists of several variants of the Ulfberht inscription, which are recorded in six cases as complete inscriptions or merely fragments thereof. The blade of a sword from Gnezdovo, Russia, features the inscription variant +VLFBERH+T (Kainov 2012: 41, Fig. 29.4), which corresponds to Type 1 in the classification by Anne Stalsberg and belongs to the most widespread groups of this inscription (Stalsberg 2008: 96, Fig. 1.1). According to Alan Williams, blades bearing this variant were made partially or entirely of steel with a higher carbon content; the researcher divided them into two groups – Group I and Group II (Williams 2012: 118, 120). These should therefore represent high-quality blades made from hypereutectoid (greater than 0.8% carbon) or eutectoid steel (approximately 0.8% carbon) (Williams 2012: 118, 120). The same variant of the inscription is also said to be present on a sword blade recovered from the waters of the Dnieper between Kherson and Oleshky (Takhtay 1935: 133; Kirpichnikov 1966a: Cat. No. 42; Stalsberg 2008: Tab. 1, Cat. No. UA1). The inscription on a blade from the Russian site of Leonovo could also be assigned to Type 1, as the letters and crosses match other inscriptions of this type, with the exception of the final letter T, which is inverted: +VLFBERH+⊥ (Kochkurkina – Summanen 2021: Fig. 1.2–4). It is assumed that an inscription in this form may indicate later production or could be an imitation originating from a workshop outside Frankish territory (Kochkurkina – Summanen 2021: 253). According to Anne Stalsberg, the inscription on a sword blade from the Aizkraukles cemetery in Latvia also belongs to the first variant (Stalsberg 2008: Tab. 1, Cat. No. LV3), which is interpreted in the form of +VLFBERHH (Müller-Wille 1970: Cat. No. 56, Fig. 7.56). A blade of a possible Type T2 sword from the Estonian island of Saaremaa bears an inscription identified as +VLFBERH+ on one side (Ebert 1914: Fig. 7; Müller-Wille 1970: Cat. No. 60, Fig. 8.60; Stalsberg 2008: Tab. 1, Cat. No. EST8). Alan Williams views this inscription as the first variant according to A. Stalsberg, i.e., +VLFBERH+(T) (Williams 2012: 139, Cat. No. II.5). According to analysis, the carbon content ranges between 0.3% and 0.5%, and despite the hypereutectoid carbon content, he places the blade in Group II (Williams 2012: 140). The inscription on a sword blade discovered at the Linkuhnen cemetery (now Rzhevskoye in the Kaliningrad Oblast, Russia) belongs to the seventh type of inscriptions according to Anne Stalsberg (Stalsberg 2008: Tab. 1, Cat. No. RUS16). The blade has not survived to the present day, but an older photograph shows that distinct letters identified as ERT – E♇T are visible on one of its fragments (Goßler – Jahn 2019: 185, 221–222, Fig. 52). In the work of Michael Müller-Wille, the inscription is supplemented by an incomplete initial Greek cross and a fragmentary letter V placed immediately after the cross. Likewise, two fragmentary marks that could have formed the letter B are also visible: +V…(B)E♇T (Müller-Wille 1970: Cat. No. 50, Fig. 7.50).
Fig. 9. a. Close-up of the blade of the sword from Leonovo with an inscription and mark (after Kochkurkina – Summanen 2021: Fig. 1.4); b. reconstruction of the inscription and mark on the blade of the sword from Leonovo (after Kirpichnikov 19192: Fig. 1.1).
The second group of inscriptions encountered within the studied type consists of two forms of Ingelrii inscriptions. In the first case, the presence of such a variant is merely presumed. According to Alexander Ruttkay, the blade of the sword from Košice – Krásna nad Hornádom allegedly bore an inscription that he identified as +INGELRII (Ruttkay 1975: 152; 1976: Fig. 28.4a). Currently, the prevailing view is that due to the condition of the sword, the inscription is illegible, even on X-ray radiographs (Môc 2018: 221; Maczek et al. 2024: 282). However, more recent research using X-ray radiographs reportedly identified the letter S and possibly O on the inscribed side of the blade, which could theoretically form a variant of the inscription +INGELRIISO (Maczek et al. 2024: 282). The second sword featuring one of the variations of this discussed group of inscriptions is the sword from Momrak, Norway (C28352). In some sources, this inscription is interpreted as +INGERIHFECIT (Kirpichnikov – Stalsberg 1998: 509, Fig. 1.3), while in others it appears as +INGERIIIFECIT (Martens – Astrup 2021: 124). Based on several detailed photographs on the Unimus portal, the latter variant appears more probable.
Compared to the Ulfberht inscriptions, this group of inscriptions is considered chronologically later; thus, these inscriptions occur on sword blades usually dated to the end of the Viking Age and thereafter (Moilanen 2015: 140).
Fig. 10. The blade of the sword from the Norwegian site of Momrak with the inscriptions +INGERIIIFECIT and CONSNVIIINS. Adapted from unimus.no.
In addition to the two aforementioned groups of inscriptions, some swords feature other variants of text. Among them is the previously discussed sword from Momrak. On one side of the blade, a variant of the Ingelrii group is visible, while on the other side, the inscription CONSNVIIINS has been identified (Kirpichnikov – Stalsberg 1998: Fig. 1.3; Moilanen 2015: 149; Maczek et al. 2024: 292). A second sword, a find from Utgården, Norway (C1648), features letters forming the inscription REX on one side of the blade, while the letters O and N were detected on the reverse (Petersen 1919: 151; Leppäaho 1952: 148; Moilanen 2015: 149). A close analogy to both inscriptions is a fragmentary blade from the Finnish site of Mynämäki (KM 8911:91), which features the variant CONSTMIITNS with mirrored letters N (COИSTMIITИS) on one side and the inscription REX on the other (Leppäaho 1952: 146–148, Fig. 1, 2; Moilanen 2015: 149, 389). Both inscriptions are interpreted as a combination of the words “Emperor Constantine”, suggesting they might represent the Byzantine Emperor Constantine VIII (976–1028); according to Jorma Leppäaho, the blade could have been part of a sword belonging to a member of the Varangian Guard (Leppäaho 1952: 149–150). The blade of the Momrak sword thus bears another version of the name Constantine (Moilanen 2015: 149), and based on the Mynämäki blade, we can assume that the two letters O and N on the blade of the Utgården sword may belong to one of the aforementioned variations or another representing the name Constantine in connection with the inscription REX, i.e., King/Emperor Constantine.
Fig. 11. a. The sword from Utgården with the inscription REX on the blade (adapted from unimus.no);
b. fragment of the blade of the sword from Linkuhnen with the preserved letters ERT (after Goßler – Jahn 2019: Fig. 52).
In addition to inscriptions, geometric marks are also recorded on Type T2 sword blades, which in most cases are combined precisely with the inscriptions. In five instances, this is a so-called lattice pattern with vertical lines – usually three – placed on both sides of the lattice (Saaremaa, Sysmä, Gnezdovo, Leonovo, Košice – Krásna nad Hornádom). An exception is a sword with a fragmentary blade from the Finnish site of Nousiainen (KM 3699:3), which is decorated with a lattice pattern on both sides (Moilanen 2015: 369).
According to N. A. Chernyshev, the blade of the sword discovered in the waters of the Dnieper during the construction of the Dneprostroi hydroelectric power station was decorated on both sides with two crosses with expanded ends (Chernyshev 1963: 223, Fig. 8). The blade of the sword from the Finnish site of Köyliö (KM 8602:130) has been preserved in a very poor condition, but despite this, inlays resembling the letters H and I were recognised on it, with two fragmentary vertical lines on the reverse side (Moilanen 2015: 387).
Apart from the organic material from the grip recorded on the blade tang, fragments of the structural components of the scabbard are also documented. Remains of wood, textile, leather, and fur were detected on the blades of only two Finnish swords—KM 8602:130 and KM 8602:141 (Tomanterä 1972: 24–26; Moilanen 2015: 387; 2018: 125).
We record only one scabbard chape, which was discovered together with a Petersen Type T2 sword in the territory of Latvia by a detectorist (Tomsons 2018: 142) (see the Find Context and Dating section below).
A noteworthy example is the sword blade from Kjølstad Vestre, Sør-Odal in Innlandet county, Norway (C37550/a), which features a so-called double or twin fuller. This means that the central part of the blade is divided by a ridge passing through its centre. Blades “decorated” in this manner are known from approximately nine or ten swords dating from the 8th to the 12th century. The origin of this blade modification is associated with the Frankish Empire, and its purpose may indicate either residues from the manufacturing process or an intentional modification of wider blades to increase the rigidity of the central section. Given that this is a relatively complex topic, we have chosen not to address it extensively; however, for those interested in a more detailed study, we highly recommend the recently published excellent work by Tomáš Vlasatý – Swords with Double Fullers: 8th–12th Centuries (Vlasatý 2026).
Fig. 12. A Petersen Type T2 sword from Kjølstad Vestre, Sør-Odal, Innlandet, Norway – C37550/a.
Adapted from unimus.no.
Decoration
As stated in the first part and also here in the introduction, in the case of Petersen Type T2 swords, we record polychrome decoration featuring a geometric pattern in the form of alternating triangles and lozenges. We have also mentioned that this pattern occurs in two variants. In addition to these, however, we also record atypical decoration within the studied type, which has been documented in only a single instance.
Geometric Pattern
Jan Petersen described the decoration of the second group of Type T swords similarly as specimens decorated with silver – much like the first variant – with the difference that the pattern consists of triangles and quadrilaterals made of silver strips arranged in a specific sequence (Petersen 1919: 151). Elsewhere, he suggests that the pattern might have been influenced by the East (Petersen 1919: 153). The ornamentation has been preserved on approximately half of the swords, in most cases only fragmentarily.
The method by which the decoration was applied to the swords of the studied type corresponds to the inlay technique. In this method, the iron surface of the pommels and crossguards is covered with vertically incised, densely spaced grooves into which non-ferrous metal wire is subsequently hammered (Vlasatý 2016a; Môc 2018: 228). The wire was typically larger in diameter than the groove, but after being hammered into the channel, it created a continuous, coloured coating (Môc 2018: 228). The density of the grooves usually varies between 10 and approximately 30 grooves per cm (Môc 2018: 228). In the case of the sword from Košice – Krásna nad Hornádom, a density of 25 grooves per cm is assumed (Môc 2018: 229). For the sword from Gnezdovo, the inlay density is estimated at approximately 30 to 45 wires per cm (Kainov 2012: 40). Two Finnish Type T2 swords feature a wire density of 23 rods per cm (Moilanen 2015: 253, 387) and 29 rods per cm (Moilanen 2015: 253, 369).
The fundamental motif of this sword type is the alternation of triangles and lozenges (rhombuses). The fields representing the triangles are filled with a white non-ferrous metal, silver. The lozenge-shaped fields between the triangles are, in most cases, composed of alternating yellow and silver strips. The yellow colour has been identified as a copper alloy (brass or bronze). This means that the rhombus field is filled with two wires of silver and brass/bronze twisted together, resembling filigree (Môc 2018: 223–224) (Fig. 15e). The triangles and lozenges are separated from each other by narrow strips of copper. This tri-colour combination has been observed on six swords (Köyliö – KM 8602:130/141; Leonovo; Košice – Krásna nad Hornádom; Dnieper – between Kherson and Oleshky, Dneprostroi). In the case of the decoration on the swords from Köyliö (KM 8602:130) and Krásna nad Hornádom, decoration made of silver, copper, and brass is presumed (Moilanen 2015: 253, 387; Môc 2018: 223–225, 230). Another four swords likewise feature hilt components decorated with silver and copper. On the second sword from the Köyliö site (KM 8602:141), Mikko Moilanen states that decoration consisting of silver and copper wires is present (Moilanen 2018: 125). However, a closer look at the decoration of the crossguard reveals certain areas filled with yellow metal (Moilanen 2018: 232, bottom-left figure). The yellow metal on the sword from Leonovo has not been precisely identified but is estimated to be merely a copper alloy, which could be either bronze or brass (Kochkurkina – Summanen 2021: 252). For both Dnieper swords, the yellow metal was initially considered to be gold (Takhtay 1953: 131–132; Chernyshev 1963: 223; Kirpichnikov 1966a: 28), though it appears more probable that it is a copper alloy (bronze or brass). Furthermore, while the sword from the Finnish site of Nousiainen does not retain its pattern, silver and brass wires have been identified on its crossguard and upper guard (Moilanen 2015: 369). It can be assumed that a part of the decoration consisting of alternating silver and brass rods, which filled the rhombus field, has been preserved on this sword. The sword originating from the destroyed barrow in Gnezdovo is covered with decoration consisting exclusively of silver and copper (Kainov 2012: 40).
On the sword from Kjølstad Vestre, Sør-Odal, Innlandet county in Norway (C37550/a), decoration made of silver and copper is preserved according to the Unimus portal; Alexandra Shchedrina, however, reports silver in combination with a copper alloy (Shchedrina 2023: 94).
Fig. 13. Hilts of decorated Type T2 swords:
a. the first variant (adapted from unimus.no);
b. the second variant (after Kirpichnikov 1966a: Pl. V.3).
We have indicated above that the pattern can be observed in two variants. The differences are noticeable on the upper guard and the crossguard, where the layout of the pattern is identical.
The first variant is composed of three sections (Fig. 13a). On the lateral sides of both hilt components, the pattern is applied horizontally. This means that triangles alternating with lozenges project from the terminal edge of the crossguard and upper guard, and another row of triangles is positioned opposite them so that their apexes point towards one another. The central section, by contrast, is filled with a vertically oriented pattern; the triangles project from the upper and lower edges of the crossguard and upper guard with their apexes pointing towards each other, thereby creating a lozenge field in the space between them. As mentioned above, the boundaries between the individual geometric shapes are filled by a narrow copper strip. We were able to distinguish the division of the composition itself on only a few specimens, depending on the state of preservation of the decoration and the quality of the available photographs. This division can be achieved either by a narrow vertical strip of copper, as seen on the swords from Košice – Krásna nad Hornádom and Köyliö – KM 8602:141 (Môc 2018: 225; Moilanen 2018: 232, bottom-left figure) (Fig. 14d, e), or by one or more lines of twisted, bi-colour wires (Utgården, Leonovo), of which, however, only the silver has survived. Both swords appear to have also had narrow copper strips applied on both sides between the band of polychrome twisted wire (Kochkurkina – Summanen 2021: Fig. 2.2) (Fig. 14a1, c1–2). In the case of the Gnezdovo sword, the division is formed by the pattern itself. The space between the horizontal and vertical decoration is filled with a narrow band of twisted silver wire, which already forms part of the vertical stepped ornamentation (Fig. 14b1).
Fig. 14. a. The sword pommel from Utgården. a1. close-up of the decoration on the upper guard of the sword from Utgården (a–a1. adapted from unimus.no); b. the sword pommel from the destroyed barrow in Gnezdovo; b1. close-up showing the division of the pattern composition (b–b1. adapted from pikabu.ru); c. the crossguard of the sword from Leonovo; c1–2. close-up showing the reconstruction of the decoration and a section of the decoration on the crossguard of the sword from Leonovo (c–c1, 2. after Kochkurkina – Summanen 2021: Fig. 2.2).
Similarly, the lower and upper edges of the crossguards and upper guards were decorated with borders, which we could distinguish on only a small number of specimens due to the state of preservation of the ornamentation and the quality of the photographs. The lower and upper edges of the crossguards and upper guards of the swords from Utgården and Leonovo were likely bordered by two smooth strips of copper, between which a band of alternating twisted wires of silver and copper alloy was applied, as evidenced by partially preserved wires (Fig. 14a1, c1). The edges of the sword from Köyliö – KM 8602:141 may have been decorated in a similar manner (Fig. 14d).
By far the best-preserved examples of these parts can be found on the sword from Krásna nad Hornádom. The lower and upper edges of the crossguard, the upper guard, and the lower edge of the pommel cap are horizontally bordered along their entire length by alternating lines of silver and copper wire bands, creating a so-called “herringbone pattern”, between which a smooth copper strip runs (Môc 2018: 225) (Fig. 15b, g). An identical motif was observed on the lateral edges of the crossguard and upper guard, running vertically along the entire length of the components, although it is less well preserved on the upper guard and on the lateral edges of the side segments (Môc 2018: 225) (Fig. 15c, d, f). The absence of oblique channels on the pommel cap meant that the division into three segments was achieved by joining several bi-colour wires side-by-side (Ag + Cu), resulting in a zig-zag pattern bordered on the outer sides by a smooth silver rod (Fig. 15b).
A herringbone pattern made of white non-ferrous metal wires was also detected on the lower edge of the crossguard of the sword from Latvia featuring the sheet-metal wrapped grip (photograph of the decoration).
Fig. 15. a. Close-up of the hilt of the sword from Košice – Krásna nad Hornádom (after Maczek et al. 2024: Fig. 3.2); b.–d. close-up of the surface decoration on the pommel; e.–f. close-up of the surface decoration on the crossguard (b.–g. SNM archives).
Within the decoration, we can also observe the ornamentation of the upper and lower sides of the crossguard near the tang and blade, as well as the underside of the upper guard. This feature is present on the sword discovered in the Dnieper between Kherson and Oleshky in Ukraine (Takhtay 1935: 132, Fig. 4.2–4) and on the Gnezdovo sword (Kainov 2012: Fig. 29.2). Based on two photographs published in the digital catalogue of the university museums of Norway, it appears that decoration is also located on the underside of the upper guard of the sword from Kjølstad Vestre (C37550/a – 1, 2). Although the specific pattern cannot be distinguished, the surface appears to be decorated with a red non-ferrous metal.
The sword from the Dnieper was first published by A. Takhtay in the monthly journal PIDO (Проблемы истории докапиталистических обществ) in 1935 (Takhtay 1935: 130–133). Anatoly N. Kirpichnikov classified it as Type T2, whereas Jiří Košta classified it as Type T2 according to Kirpichnikov, or as a possible Type T, or alternatively Type V (Kirpichnikov 1966a: Cat. No. 42; Košta 2021: Appendix, Tab. 10, Cat. No. 3158). A research team consisting of Maczek, Môc, Dvorák, and Beronská also inclines towards its classification as an uncertain Type T2, arguing that not a single photograph or drawing of the sword exists in the literature (Maczek et al. 2024: 294). In light of the publication of the find by A. Takhtay, however, we can state that the sword unequivocally belongs to Petersen Type T2 (Fig. 16). The pattern on the crossguard and upper guard consists of alternating triangles and lozenges positioned vertically in the centre and horizontally on the sides. This variously arranged geometric pattern is also present on the undersides of the crossguard and upper guard, as well as on the upper side of the crossguard.
Fig. 16. A Type T2 sword from the Dnieper between Kherson and Oleshky (after Takhtay 1935: Figs. 1, 2, 4).
The Gnezdovo sword is of particular interest within this type due to its decoration. The composition present on the primary facing sides of the crossguard and upper guard corresponds to our first variant – vertically in the centre and horizontally on the sides (Fig. 17). However, the triangles and lozenges are not depicted in the manner customary for this type; instead, they are stepped, a feature typical of Petersen Type V swords (Petersen 1919: 155; Viskupič 2025). For this reason, it was assigned to Type V by Kirpichnikov (1966a: Cat. No. 62). Sergey Kainov disagrees with this classification, arguing that given the morphology of the hilt components and the composition of the decoration, the sword bears characteristics distinctive to Type T2 (Kainov 2012: 39, Note 28). Other researchers view the sword as a specimen with mixed elements of both Types T2 and V (Maczek et al. 2024: 294).
The stepped triangles are bordered by a narrow copper strip, and the internal fields of the lozenges are filled with twisted silver and copper wires. This style of decoration is typical of many Type V swords (Jahn – Ibsen 2024: Fig. 18, 20; Viskupič 2025), but it can also be encountered on Types H/I. Two such swords originate from Sweden – Norra Kvinge, Öland (SHM 16646) and Birka, grave 561 (SHM 34000:561) and a third from Norway – Nord-Aurdal, Innlandet (C23127). Regarding Type H/I, Alfred Geibig adds that on swords of his combinational Type 5 (Petersen’s Type H/I), this method of decoration is less common (Geibig 1999: 37). It may further be added that on Types H/I and V, stepped triangles with lozenges are always applied vertically. The underside of the upper guard is decorated with an identical motif to that found on the facing parts of both components (the upper guard and crossguard); the lateral sides, at the location of the rivets, are decorated with vertical bands, likely consisting of alternating twisted silver and copper wires. The upper side of the crossguard is again decorated with silver and copper – the centre around the slot for the tang is filled with a stepped pattern, while the lateral sections are covered with semi-circular and circular ornaments. The terminal edges are presumably decorated with silver and copper bands. The reverse side is largely decorated with alternating bands resembling a zig-zag pattern (Ag + Cu) and the ends with copper shapes on a silver background in the form of intertwined lines (Kainov 2012: 40).
Fig. 17. A Type T2 sword from the destroyed barrow in Gnezdovo (after Kainov 2012: Figs. 29.2–3, 30.1–2).
Based on our observations, the pattern of our first variant is present on eight specimens (Köyliö – KM 8602:141, Latvia – Sigulda region, Utgården, Kjølstad Vestre, Gnezdovo, Leonovo, Košice – Krásna nad Hornádom, Dnieper – between Kherson and Oleshky).
Finally, a sword originating from the Lithuanian site of Viešvilė, Jurbarkas district, near the border with the Kaliningrad Oblast, should be mentioned. According to a drawing published in the work of Regina Volkaitė-Kulikauskienė, the decoration is allegedly preserved on the upper guard and crossguard. This ornamentation consists of silver decoration made of straight lines – two bordering the upper and lower edges of both guards along their entire length; approximately at the position of the blade tang, two vertical bands are applied, filled in the centre with two short horizontal lines; on the lateral sides, there is a pattern resembling an oblong rectangle through the centre of which a single horizontal line passes (Volkaitė-Kulikauskienė 1964: 169, Fig. 12). It is not clear from the illustration what specific pattern this sword features, nor what condition the decoration is currently in; however, based on the morphology of the pommel and crossguard, the researcher classified it under Type T (Volkaitė-Kulikauskienė 1964: 225). Ugnius Budvydas agrees with this classification (Budvydas 2007: 210). Some researchers consider the sword to be a Type V (Kazakevičius 1996: 115, Cat. No. 12; Janionis 2025: 333, Cat. No. 139), a view with which one cannot agree, since the pommel exhibits morphological features that unequivocally link it with Type T, more precisely the T2 subtype.
Fig. 18. A Type T2 sword from Viešvilė:
a. close-up of the decorated hilt (after Volkaitë-Kulikauskienë 1964: Fig. 12);
b. current condition of the sword (after Janionis 2025: 333).
The second pattern is characterised by alternating triangles and lozenges formed by non-ferrous coloured wires. The composition, however, is different; the decoration is arranged vertically along the entire length of the crossguards and upper guards.
We have identified this pattern on seven specimens (Köyliö – KM 8602:130, Dneprostroi, Øvrebø, Vakhrushevo, Chernihiv – Black Grave, the Lower Dvina region, Upa). The tri-colour combination – silver triangles with lozenges filled with alternating twisted silver and brass wires, separated from one another by a copper contour – is recorded exclusively on the sword from Köyliö – KM 8602:130 (Moilanen 2015: 387). Polychrome decoration was also preserved on the sword from Dneprostroi, where, as noted above, the yellow metal was initially considered an alloy of gold. Given that this sword, along with four other specimens discovered at the same site, was lost from the Dnipropetrovsk Museum during the Second World War, a more detailed analysis is currently impossible. However, it is evident from surviving photographs that the internal fields of the lozenges are filled with alternating twisted wires (Komar 2014: Figs. 8.9–10).
Fig. 19. The sword from Dneprostroi:
a. view of the complete sword (after Komar 2014: Fig. 5.4);
b. close-up of the hilt (after Kirpichnikov 1966a: Pl. V.3);
c1. close-up of the decoration on the pommel (after Komar 2014: Fig. 8.9);
c2. close-up of the decoration on the crossguard (after Komar 2014: Fig. 8.10).
The sword from Øvrebø, Norway (C17958), has been preserved with a damaged blade and crossguard, on which the remnants of decoration are discernible in the form of a small number of triangles on both sides. In addition, decoration consisting of twisted silver wires has survived in places. The description by the Museum of Cultural History in Oslo on the Unimus portal states that the crossguards are similar in shape to the R494 specimens (the equivalent of Petersen Type H/I) and feature a pommel with a three-part division and side edges stylised as animal heads. Photographs, as well as a drawing with a reconstruction of the vertical decoration of alternating triangles and lozenges, clearly show that the sword is currently missing its pommel cap (if indeed it was ever found with this component), a point also highlighted by other researchers (Maczek et al. 2024: 292). Jan Petersen likewise offers no detailed information regarding the condition of this sword. Elsewhere, it is viewed as a possible Type T or V (Martens – Astrup 2021: 145, Appendix 1).
Fig. 20. Close-up of the crossguard and general view of the sword from Øvrebø.
Adapted from unimus.no.
Decoration using silver and copper-alloy wires is also documented on the sword from Vakhrushevo in the south-eastern Lake Ladoga region (Kirpichnikov 1966a: Cat. No. 39; Shchedrina 2021: 272). It is unclear whether this sword has the internal fields of its lozenges filled with contrasting twisted wires. Based on the morphology of the hilt’s structural parts, Alexandra Shchedrina compares the sword from Vakhrushevo and the Norwegian specimen from Utgården with another Type T2 sword discovered at the Shuya site, which, however, was preserved without decoration (Shchedrina 2021: 272, Figs. 1, 2.1–3). The central segment is more semi-circular and the lateral lobes are not prominently concave, making the pommel closer to Type V. Some researchers view this sword as a hybrid form or a transitional variant from Type V to Type T2 (Maczek et al. 2024: 294).
Another sword with this type of decoration originates from the Black Grave in Chernihiv, which survives only with its hilt and a small portion of the blade. The material of the decoration is unknown, but based on two older black-and-white photographs available in the specialist literature, vertically arranged triangles and lozenges can be seen along the entire length of the crossguard (Samokvasov 1916: Fig. 25; Androshchuk – Zotsenko 2012: Fig. 222). This sword is regarded by some researchers as a Petersen Type T2 (Kirpichnikov 1966a: Cat. No. 41; Androshchuk – Zotsenko 2012: 288). Others, by contrast, view it as a possible Type T or V, or as an uncertain Type T2 (Košta 2021: Tab. 10, Cat. No. 3163; Maczek et al. 2024: 294). Based on the morphological attributes of the hilt, Sergey Kainov inclines towards Type T2 (personal communication with Sergey Kainov). Only one Type T2 sword with this type of decoration is known from the territory of Latvia. The sword reportedly originates from illicit detectorist activity and was discovered somewhere along the lower reaches of the Dvina River; it is currently held in a private collection (Tomsons 2018: 62–63, Fig. 29.9). The final sword known to us on which we have identified the second variant of decoration originates from the Estonian site of Upa, discovered at the Kalmumäe cemetery (Mandel 1991: 117, Fig. VIII.1). Mati Mandel states that the decoration has survived in the form of lozenges made of silver inlay. From the available photograph, which is not of the highest quality, it can be concluded that vertically succeeding triangles are visible in places around the entire perimeter of the crossguard and upper guard. On the pommel cap, the decoration appears to consist of a white non-ferrous metal in the form of triangles and lozenges.
Fig. 21. a. Köyliö – KM 8602:130, Finland (after Moilanen 2018: 119);
b. Vakhrushevo, Russia (after Shchedrina 2021: Fig. 2.1);
c. Chernihiv – Black Grave, Ukraine (after Androshchuk – Zotsenko 2012: Fig. 222);
d. the lower reaches of the Dvina, Latvia (after Tomsons 2018: Fig. 29.9);
e. Upa, Estonia (after Mandel 1991: Pl. VIII.1).
Based on our observations, we infer that none of the swords decorated with vertical geometric shapes likely possess decoration on the underside of the upper guard, nor on the crossguard near the tang and blade. From the photographs available to us, we assume that decoration was present on the upper and lower edges of the crossguard of the sword from Köyliö – KM 8602:130 (Moilanen 2018: 119) and on the detector find from the lower Dvina, as evidenced by close-up shots of the hilt’s structural components published on the antik-war.lv online forum. Decorative borders around the edges were also likely present on the sword from Dneprostroi (Fig. 19c1–2). According to Mikko Moilanen, the sword from the Finnish site of Sysmä, Haapasaari – KM 31017:1 (Moilanen 2015: 417) allegedly features hilt components decorated similarly to the sword from Köyliö – KM 8602:130, including the arrangement of the grooves, which are said to be applied in two directions (Moilanen 2015: 253–254).
Fig. 22. A Type T2 sword from Sysmä, Haapasaari, Finland – KM 31017:1 (after Moilanen 2015: 418).
In our estimation, a sword originating from an unknown Swedish site could also belong to the type under study. It is housed in the collections of the Swedish History Museum in Stockholm (Statens historiska museer – SHM), where it is recorded only under the item number 266352_HST. Apart from the museum’s digital catalogue, the sword has not been extensively published. It was likely first mentioned by Sue Brunning (2015) and subsequently appeared in two articles published online (Vlasatý 2022b; Viskupič 2025). The most recent mention in a printed periodical dates to 2025 (Kainov – Shchedrina 2025: 233).
This is a fragmentary sword preserved only with a portion of the blade and a hilt lacking its pommel cap (Fig. 23a). The decoration is best preserved on the crossguard. It consists of vertically arranged white triangles alternating with yellow lozenges, between which a red contour is applied. According to information from the SHM portal, the decoration was executed using brass, copper, and silver. The internal field of the lozenges, however, lacks any infill of mixed twisted wires; that is to say, the fields are covered with a solid layer of brass. Decoration is also applied to the underside of the upper guard, where a pattern of alternating silver and brass triangles with copper bordering can be observed in the section near the tang. The remaining parts likely featured a pattern identical to that on the crossguard and the front facing side of the upper guard, though the motif is more poorly preserved there. The underside of the crossguard is likewise covered with a polychrome geometric pattern of alternating triangles and lozenges in a stepped pattern within its narrow sections; towards the centre, the pattern transitions into alternating smooth lines of silver, copper, and brass (historiska.se).
The stepped pattern, as mentioned above, is primarily characteristic of Type V swords. This pattern is recorded in two variants—filled exclusively with silver and copper/copper alloy (Viskupič 2025: Fig. 12) or tri-colour, utilising silver, copper, and brass/bronze (Viskupič 2025: Fig. 13). Typologically, the sword is interpreted in different ways. Sue Brunning and Fedir Androshchuk consider it a later variant of Petersen Type H/I (Brunning 2015: 61–62; Vlasatý 2022b). At the same time, F. Androshchuk states that not a single Type T sword originates from the territory of Sweden, with the exception of the unfinished Type T1 upper guard from Gotland (Androshchuk 2014: 77; Viskupič 2026a). Sergey Kainov and Alexandra Shchedrina regard it as a Type V (Kainov – Shchedrina 2025: 233). In our view, we consider the sword a possible Type T2 or Type V (Vlasatý 2022b; Viskupič 2025). The absence of the pommel cap prevents us from drawing more precise conclusions. In its shape, the pattern resembles the decoration of Type T2 swords, more specifically according to our second variant. On the other hand, the missing twisted-wire infill of the lozenges suggests a proximity to the decoration seen on some Type V specimens. We cannot rule out the possibility that the weapon represents a transitional variant from Type V to T2, or vice versa.
Fig. 23. a. A sword of an uncertain Type T2 or V from an unknown Swedish site (adapted from historiska.se); diagram showing the decorative layout of the crossguard and upper guard with the placement of pommel caps of the following types: b. Petersen H/I, c. Petersen V, d. Petersen T2 (drawing by author).
Let us also briefly examine the decoration of an independent pommel cap. The central segment is decorated in a manner similar to that seen on Type V swords (Viskupič 2025: Fig. 17). The motif consists of horizontally arranged silver triangles bordered by a copper strip. Higher up, towards the apex, brass lozenges (filled with alternating silver and brass wires) alternate with silver ones. On most swords, the pattern is preserved only fragmentarily and can be documented on eleven specimens (Upa, Köyliö – KM 8602:130; Sysmä; Curonia – Latvia; Kjølstad Vestre; Utgården, Gnezdovo, Vakhrushevo, Košice – Krásna nad Hornádom; Dnieper between Kherson and Oleshky, Dneprostroi).
Fig. 24. The pommel cap of a sword discovered on the lower reaches of the Venta River, Curonia, Latvia.
Private collection. After Tomsons 2018: Fig. 30.1.
The decoration of the lateral segments could be identified on only seven swords (Köyliö – KM 8602:130, Curonia – Latvia, Kjølstad Vestre, Utgården, Gnezdovo, Košice – Krásna nad Hornádom, Dnieper between Kherson and Oleshky, Dneprostroi). The pattern on most of them has survived in a more or less fragmentary state, with the exception of the Dnieper find, where it is documented only in the form of a drawing and the actual condition of the decoration is unknown to us (Takhtay 1935: Fig. 4.5). Despite the incomplete preservation of the motif on the swords, it can be stated that this pattern exhibits variations. This indicates that the lateral segments were likely covered with decoration that differed in its details. The basic pattern is shaped into the form of a silver triangle, which is alternately bordered by larger triangles made of thinner copper and silver strips, or by a thicker contour of copper strips (Môc 2018: Fig. 6c; Kainov 2012: Fig. 29) (Fig. 15b–c; Fig. 17). The sword from the Finnish site of Köyliö – KM 8602:130 probably has the field of the triangles filled with copper wire (Moilanen 2018: 119). Beneath the triangles run obliquely positioned bands, also consisting of alternating silver and copper wires. Vladimir Kulakov identified this symbol as flags, or alternatively, its meaning is interpreted as a stylised form of eyes with the bared teeth of a beast (Kulakov 1989: 69, Fig. 4.3; Môc 2018: 224). The remaining parts of the segments exhibit different forms of decoration: the Gnezdovo sword has the “flags” surrounded by a stepped pattern on a silver background with a copper contour and an internal field composed of twisted copper and silver wires (Kainov 2012: 40, Fig. 29); on the sword from Kjølstad Vestre, Norway, we record alternating smooth bands of copper and silver (observation based on the photographic documentation of the Unimus portal); on the Košice specimen, the remaining part of the decoration is more or less completed by a zig-zag pattern that optically separates the central segment from the lateral ones; a torso of silver wires is preserved on the lateral segments of the sword from Utgården, Norway, and on the find of an independent pommel cap from an unspecified location on the lower reaches of the Venta River in the historical region of Courland in Latvia (observation based on the photographic documentation of the Unimus portal; Tomsons 2018: Fig. 30.1). A similar characteristic can be observed on Type V swords, where we have managed to identify approximately fifteen specimens that have survived with decoration on these parts. The basic shape is formed by a triangle-like motif, which is intersected in its widest part by an oblique vertical line of red non-ferrous metal wire, while the remaining parts of the decoration differ on individual specimens (Viskupič 2025: Fig. 17).
Fig. 25. Examples of preserved decoration on the lateral segments of Petersen Type T2 sword pommel caps:
a. Gnezdovo – destroyed barrow (after Kainov 2012: Fig. 29); b. Košice – Krásna nad Hornádom (SNM archives).
Finally, two swords of a possible Type T2 from Norway should be mentioned. One of them has already been noted in the literature; it is a sword find with a fragmentary blade from Mårem Nordre, Tinn in Telemark county (C52806/1) (Maczek et al. 2024: 292). The sword is a grave find and is heavily corroded. The description on the Unimus portal classifies the sword as a Mannheim type, stating that its close analogy is the specimen originating from the Norwegian site of Elvran Øvre, Stjørdal, Trøndelag county (T18969), or alternatively, its classification is given under Petersen’s Special Type 1 – Særtrype 1 (Martens 2009: 186). In our opinion, the condition of the pommel does not allow for a more precise typological assignment; nevertheless, its division into an upper guard and a three-part pommel cap with fragmentary decoration, probably of silver, is discernible. The lower part of the upper guard is covered in some places with small strips of white non-ferrous metal. The upper side of the upper guard features a clearly recognisable triangle symbol made of white non-ferrous metal on one side. In addition, one of the lateral segments retains oblique bands in its lower section, presumably silver, which may be the remnants of decoration representing Kulakov’s flags/bared teeth of a beast.
Fig. 26. A sword of a possible Type T2 from Mårem Nordre – C52806/1 (adapted from unimus.no).
The second specimen has likely not been published yet. This grave find from Sømhovd, Sømna in Nordland county, northern Norway (C13815), was preserved in a very poor condition and consists of approximately eleven fragments. The sword is housed in the collections of the Museum of Cultural History in Oslo, and according to the description on the Unimus portal, the hilt is similar to a Type R494 sword. This type according to O. Rygh (1885), as mentioned above, belongs to Type H/I within Petersen’s typology (Petersen 1919: 89). Both crossguards are straight in frontal view, with a lenticular plan view; however, the pommel cap is not three-part but triangular (Petersen 1919: Figs. 79, 82, 86–87). Based on the photographs published on the Unimus portal, however, we can state that the sword is more likely to belong to Type T2. Our argument rests on several observations. Although the pommel is not preserved in the best condition, we can nevertheless recognise a lateral segment with a characteristic concave section and a relatively high ridge. Furthermore, twisted wire (presumably silver) arranged obliquely in three parallel rows has been preserved in this section. This element is not found on Type H/I swords and is typical primarily for the three-lobed pommels of Type E, T, and V swords, but also for Type R, S, and Z swords with multi-lobed pommel caps.
Another reason why we believe the sword may belong to the type under study is the partially preserved decoration, presumably of silver. This is relatively clearly visible on one side of the pommel cap in the form of a triangle projecting from its lower side and a symbol resembling a lozenge. In addition, another white non-ferrous metal triangle is preserved on the upper side of the upper guard. Decoration resembling triangles was also observed in two places on the crossguard. Given the condition of the sword, we cannot unequivocally confirm its assignment to Type T2, but the aspects described above allow us to designate it as an uncertain Type T2.
Fig. 27. A sword of a possible Type T2 from Sømhovd, Sømna, Nordland, Norway – C13815 (adapted from the unimus.no portal). A close-up view is available via this link.
Atypical Pattern
In addition to the geometric pattern of alternating triangles and lozenges, we also observe an atypical decoration on this type, consisting of an interlaced motif. We find it on the sword from grave 53/1929 (PM Pr 303b) from the East Prussian cemetery at Linkuhnen (now Rzhevskoye, Kaliningrad Oblast), of which only the two-part pommel and crossguard have survived to the present day (Mühlen 1975: 33, Fig. 10.1; Kazakevičius 1996: Cat. No. T-50; Goßler – Jahn 2019: 185–189). The blade of the sword was described above, so we will not address it here. We will focus exclusively on the decoration and morphology of the structural components.
The decoration has not survived intact; silver lines of intertwined loops and dots are discernible. Based on the originally good state of preservation, Wolfgang La Baume attempted to analyse the technique of applying the decoration and to produce a reconstruction of the motif (Goßler – Jahn 2019: 185).
The pattern on the upper guard and crossguard is more or less identical. Thin silver strips in the form of intertwined loops with four stylised square-shaped knots run along the entire length of the crossguard; the space of the upper guard is filled in a similar manner, with the difference that the loop strips are elongated, there are only three square knots, and both ends of the upper guard are filled with a horizontally positioned motif resembling the letter T, from which two strips branch out to the sides, creating a pattern similar to the letter Y (Goßler – Jahn 2019: 185–186, Figs. 52, 53). The three-lobed pommel cap was probably divided by a zig-zag pattern (Goßler – Jahn 2019: 185, Fig. 53). The central segment is decorated with looped, coiled bands that split towards the lower edge of the pommel cap, creating a pattern resembling an axe head. The decoration of the lateral lobes consisted of double lines of silver bands following their shape. The background pattern was supposedly created from irregularly and densely arranged circular particles.
According to W. La Baume, the decoration was executed using the plating technique; that is, the substrate was first roughened with parallel and grid-like incised grooves and depressions onto which a solid silver sheet was to be fastened, from which the lines of the ornament were subsequently cut out (Goßler – Jahn 2019: 186). Vladimir Kulakov also states that the decoration was created by the plating technique (Kulakov 2019: 75). Norbert Goßler and Christoph Jahn consider such a method unlikely, inclining instead towards inlaid decoration made of twisted silver wires placed into pre-prepared grooves, as well as in the case of the flat dotted decoration, which took the form of small silver plates inserted into pits (Goßler – Jahn 2019: 186). From the photograph of the pommel available in the digital catalogue of the State Museums in Berlin (Staatliche Museen zu Berlin), vertical, oblique, or grid-like incisions can be observed on the iron surface upon closer inspection. It appears that the grooves do not correspond with the motif.
Fig. 28. The pommel and crossguard of the Type T2 sword from grave 53/1929 at Linkuhnen – PM Pr 303b. Adapted from recherche.smb.museum.
Vladimir Kulakov considers the decoration to be close to the Mammen style, and in his view, the trapezoidal shapes on the lateral parts of the pommel cap may be stylised animal or bird heads, while the central lobe is said to bear the form of a headless bird (Kulakov 2019: 75).
The composition of intertwined lines with knots strikingly resembles the decoration of certain Type S swords. The closest analogy is a sword with a bent blade originating from Århus, Fyresdal, Telemark county – C11451 (Müller-Wille 1972: Cat. No. 15, Fig. 30.2). According to information from the Museum of Cultural History in Oslo, the hilt components were covered with silver inlay in the form of a braid ornament. The silver later melted, causing it to accumulate into clusters of droplets. The photographs of the sword show that the decoration was applied into pre-prepared incisions on the iron surface. In this case, however, the incisions correspond to the motif, meaning that grooves were cut into the iron surface in the intended pattern (Môc 2018: 227). In addition to the Norwegian specimen, similar decoration is documented on two other Type S swords – one was discovered on the shore of Lake Drevviken in Norrby, Österhaninge parish, Södermanland county in Sweden (SHM 20382) (Androshchuk 2014: Cat. No. Sö 11*; Wikell 2019: 3; photograph of the sword by Peter Johnsson) and the other is reportedly a find from Schwedt near the Oder River in the federal state of Brandenburg, Germany (If 16773) (Petri 2017: 158–166, Fig. 16a–b). This style of decoration can also be encountered, for instance, on Type O, R, or Z swords (Môc 2018: 228). The silver dotted decoration on the background of the interlaced motif is also identical, as can be observed on certain swords, such as one Type O (C13458) and two Type R specimens (T20362/1), all from Norway, or on a Type Z sword from Sweden (SHM 20981).
Fig. 29. a1–2. A Petersen Type S sword from Århus, Fyresdal, Telemark county, Norway and its drawing (C11451);
b. drawing of a Petersen Type O sword from Allum, Larvik, Vestfold, Norway (C13458);
c. a Petersen Type R sword from Egge, Steinkjer, Trøndelag, Norway (T20362/1);
d. a Petersen Type Z sword from Södertälje, Södermanland province, Sweden (SHM 20981).
a.–c. adapted from unimus.no; d. adapted from historiska.se.
The sword from Linkuhnen has been classified in several works under Type T1 (Goßler – Jahn 2019: 186; Košta 2021: Tab. 10, Cat. No. 2399) or as T2 (Mühlen 1975: 33). Based on the morphological attributes of the crossguard and pommel, we incline towards the possibility that the sword is closer to Type T2 rather than T1. Both components lack the massive proportions typical of the T1 variant. On the contrary, very close morphological analogies for the pommel originate from Viešvilė and Krásna nad Hornádom. They are likewise linked by the absence of grooves dividing the pommel cap and the filling of this space with silver wire in a zig-zag pattern. On one side of the lower part of the upper guard, a preserved circular hole is clearly visible, which undoubtedly served to thread the rivet connecting the pommel cap to the upper guard.
Based on these elements, we can state that the pommel and crossguard were parts of a sword that morphologically fits Type T2, although its decoration differs from the classic type as defined by Jan Petersen. The artisan was undoubtedly inspired by ornamentation derived from Petersen Type S or Type R swords.
Fig. 30. Schematic layout of the decoration on Type T2 swords: a. the first variant; b. the second variant (drawing by author).
In addition to Petersen Type T2 swords, the motif of alternating triangles and lozenges can also be encountered on certain types of spearheads, much like in the case of Type V swords, where the pattern of alternating stepped triangles and lozenges is present on some spearheads of Petersen Types E, G, I, and K (Viskupič 2025). The motif decorating the swords of the group under study is recorded on several spearheads of Petersen Type I. A perfect example of such decoration is the specimen originating from Gjermundbu, Norway (C27317/d), as evidenced by the photographs and reconstruction available on the Unimus portal. Another example of this decoration can be observed on two spearheads from Gotland, one of which comes from the site of Akebäck (SHM 16822) and the other from Dalhem (SHM 5352) (Thunmark-Nylén 2006: Pl. 243.1, 2a–c).
Distribution
Regarding their distribution, Petersen Type T2 swords exhibit a relatively higher frequency of occurrence compared to the first variant; nevertheless, they do not constitute a particularly numerous group. Within Northern Europe, they are present only in Norway and Finland. The greatest distribution of this type is recorded in Eastern Europe, including the Baltic region. A unique phenomenon is the isolated find from the territory of Slovakia, as it currently represents the only known sword of this type in Central Europe.
From the territory of Norway, we record only four swords of certain Type T2 and two specimens whose assignment to the type remains uncertain. Almost all of them are concentrated in the south of the country, within Telemark county. Two specimens originating from Utgården, Seljord (C1648) and from Øvrebø, Bø parish (Midt-Telemark) (C17958) were recorded by Petersen as the only finds of this type in Norway (Petersen 1919: 150–151). The other two are finds from Momrak, Fyresdal (C28352/a) and from Mårem Nordre, Tinn (C52806/1). Given its condition, the latter has been classified as an uncertain type. The remaining two swords were found in different parts of the country. One was recovered in the east, in Innlandet county (Kjølstad Vestre, Sør-Odal – C37550/a), and the other, which we have likewise designated as an uncertain Type T2, appeared in northern Norway, in Nordland county (Sømhovd, Sømna – C13815).
According to the Unimus portal, three additional swords and their structural components belong to Type T, which we have previously addressed in a work focused on Petersen Type V swords (Viskupič 2025). These are specimens from the counties of Telemark (Loftsgården, Seljord – C17404), Trøndelag (Fjermstad, Trondheim – T18931), and Nordland (Nesna – T14487). In our opinion, the pommel of the sword from Trøndelag exhibits features more characteristic of Type V rather than Type T. For the remaining two specimens (C17404 and T14487), we can retrospectively re-evaluate their classification based on morphological attributes and incline towards the view that the indistinct styling of the lateral segments into the shape of animal heads and the oval edge of the central segment are more reminiscent of Type V.
Fig. 31. A Petersen Type T2 sword from Utgården, Seljord, Telemark – C1648 (adapted from unimus.no).
In Finland, Mikko Moilanen records only four Type T2 swords. Two originate from the Vanhakartano cemetery at the Köyliö site—KM 8602:130 (Moilanen 2015: 387; 2018: 124) and KM 8602:141 (Moilanen 2018: 125) in the west of the country. The third sword comes from south-western Finland, from the site of Nousiainen, Kärmälä – KM 3699:3 (Moilanen 2015: 369). The fourth find is a fragmentary sword discovered by a metal detector at the site of Sysmä, Haapasaari – KM 31017:1 (Moilanen 2015: 418).
In our estimation, yet another sword from Finland could potentially belong to one of the variants of Type T, at least judging by the morphology of the hilt’s structural parts. This is a completely preserved sword discovered at the site of Loppi, Pietilä, Kanta-Häme (KM 41288:14), which according to the finna.fi portal belongs to Type S. In the photographs on the portal, the sword appears to be in its post-excavation state. Since its current condition is unknown to us, we must rely solely on the description of the shape of the crossguard and pommel. The crossguard likely features two straight arms, their ends visibly sloped obliquely towards the blade. It cannot be determined whether the pommel is formed of two components or only one, but a relatively massive central segment shaped like a triangle with a rounded apex is distinct. The lateral segments appear to be formed into the shape of animal heads with a concave transition. The decoration is not recognisable beneath the layer of soil. The assignment of the sword to Type T is therefore tentative in this case. Its affiliation with another group of swords could also be considered. The pommel and crossguard may also be close to specimens from Padasjoki in Finland and Novgorod in Russia (Viskupič 2026a: Figs. a1–2, b). These are characterised by a five-part division of the pommel cap and are typologically situated somewhere between Types S/Z and the so-called silver-plated type.
Fig. 32. a. A sword of Petersen Type T2 from Nousiainen, Kärmälä – KM 3699:3 (after Moilanen 2018: 121);
b. a Type T2 sword from Köyliö, Vanhakartano – KM 8602:141 (after Moilanen 2018: 125).
From the territory of Estonia, only two Petersen Type T2 swords are known, both specimens having been discovered on the island of Saaremaa. One originates from the Kalmumäe cemetery in the village of Upa in the south of the island (Mandel 1991: 117, Cat. No. 18) and the other from an unspecified location, which has been assigned different typological classifications in the literature. According to Mati Mandel, the sword belongs under Petersen Type V (Mandel 1991: 118, Cat. No. 23). Other authors classify it as Type T (Müller-Wille 1972: 86, Cat. No. 60; Stalsberg 2008: Cat. No. EST8). The decoration on the sword has likely not survived; therefore, a typological assignment can be attempted based on the morphology of the pommel, which according to A. Williams was allegedly lost (Williams 2012: 139, Cat. No. II/5). Although we do not know the current state of the sword, not only an older drawing but also several photographs are available (Ebert 1914: Fig. 6; Mandel 1991: Fig. X.1; Mägi 2017: Fig. 4.13). The external segments are not as prominently concave as is customary for many Type T2 swords, but in shape, they are similar to those featured, for instance, on the pommel of the sword from Vakhrushevo, Russia (Fig. 21b). The central segment, however, is higher compared to Type V swords, bringing the pommel closer to Type T2 (Viskupič 2025).
Fig. 33. A sword of an uncertain Type T2 from the island of Saaremaa – AI K85:120 (after Mägi 2017: Fig. 4.13).
In Latvia, according to Artūrs Tomsons, eleven swords of the group under study occur (Tomsons 2018: 62–65, Figs. 29, 30, Tab. 2, Cat. Nos. 41–51). Some researchers disagree with this number, suggesting that based on morphological and decorative attributes, only two swords can be assigned to this type – Cat. Nos. 50 and 51 according to Tomsons (Maczek et al. 2024: 293). Other scholars state that Type T2 swords do not appear around the Baltic Sea at all (Martens – Astrup 2021: 31). Based on our research, we cannot agree with this assertion. We believe that six specimens can be assigned to the monitored group of swords according to the morphological features of the pommel and crossguard, as well as the decorative elements. These are: the sword originating from the Aizkraukles cemetery – DM I 2899 (Tomsons 2018: Tab. 2, Cat. No. 47). Although the sword has been preserved without decoration, the shape of the pommel exhibits features characteristic of Type T2, which can best be observed in photographs on the portal of the Latvian Archaeological Society (Latvijas Arheologu biedrība). The sword is listed there under inventory number RDMI 2956, but the photographs of the object from the LAB portal match the one presented by A. Tomsons (Tomsons 2018: Fig. 29.4). On one of the photographs from the LAB portal, it appears that a small textile fragment has been preserved on the crossguard, and at the same time, a kind of wedge protruding from the crossguard near the blade tang is relatively clearly visible. Another sword that can be classified as T2 due to the characteristic shape of its pommel is the one originating from the Doles Vampenieši I cemetery – VI 124:1270 (Tomsons 2018: 62, Fig. 29.2, Tab. 2, Cat. No. 42). The pommel cap of a Type T2 sword likely originates from the historical region of Livonia; it is a detector find and is said to be part of a private collection (Tomsons 2018: 64, Fig. 30.2, Tab. 2, Cat. No. 49). Another Type T2 sword with the decoration of our second variant originates from Livonia or Curonia, somewhere in the area of the lower reaches of the Dvina, and according to A. Tomsons, it is a detector find that may come from a cremation grave (Tomsons 2018: 62–63, 142, Fig. 29.9, Tab. 2, Cat. No. 50). The fifth specimen is a pommel cap discovered on the lower reaches of the Venta River in Courland (Tomsons 2018: 64, Fig. 30.1, Tab. 2, Cat. No. 51). The sixth find is a sword originating from an unknown location in the Sigulda region (information provided to us by the finder). This is a completely preserved sword featuring the decoration of our first variant. Information regarding the find is available from the period between 2012 and 2014 on the swordmaster.org online forum.
Fig. 34. Petersen Type T2 swords from Latvia: a. Aizkraukles (adapted from arheologubiedriba.lv);
b. Doles Vampenieši I – VI 124:1270 (after Tomsons 2018: Fig. 29.2);
c. pommel cap from Livonia(?) (after Tomsons 2018: Fig. 30.2).
In addition to the five aforementioned swords that A. Tomsons classifies under Type T2, he lists another six swords (Tomsons 2018: Tab. 2, Cat. Nos. 41, 43, 44, 45, 46, 48). Although the structural parts of the swords under Cat. Nos. 43 and 46 – from Gramzdas Upītes and Kokneses kapulauks (Tomsons 2018: Fig. 29.6, 8) – exhibit features similar to the type under study, we are unable to determine from the photographs whether both specimens bear lozenge decoration. Neither of these swords retains a pommel cap; thus, we cannot rule out the possibility that they might have carried this component belonging to Types H/I or V. Furthermore, the sword under Cat. No. 43 from Gramzdas Upītes is equipped with a narrow upper guard that is slightly curved upwards. Due to the poor quality of the photograph, the sword under Cat. No. 48 – from Turaidas Pūteļi (Tomsons 2018: Fig. 29.5, Tab. 2, Cat. No. 48) – is difficult to assign to a specific type. The decoration has not survived on the other three swords either (Tomsons 2018: Fig. 29.1, 3, 7), but the morphological attributes of the pommels appear close to the type under study; based on this, we classify them as an uncertain Type T2.
Two additional specimens, assigned by A. Tomsons to the first variant of Type T – the T I subtype (Tomsons 2018: Tab. 2, Cat. Nos. 27, 28) (Fig. 35a–b) – can be considered a possible Type T2. These are a sword from the Doles Vampenieši cemetery, grave 170 – VI 124:1758 (Tomsons 2018: Fig. 25.1) and a pommel cap from the Salaspils Laukskola cemetery, grave 300 – VI 128:3405 (Tomsons 2018: Fig. 25.4). Based on our observations, we conclude that morphologically, the structural parts of the hilt of both finds are closer to Type T2. The sword under Cat. No. 27, preserved with a crossguard and a two-part pommel, likely exhibits traces of decoration made of white non-ferrous metal in the form of triangles placed vertically in the centre and horizontally on the sides of both guards. When viewing the pommel cap, silver(?) shapes resembling triangles and lozenges can be observed. The pommel cap from Salaspils Laukskola appears to retain no decoration; however, in shape it is similar to the pommel of the sword from Utgården, even though one of the better-preserved segments does not show a prominent inward indentation. At the same time, a long rivet protrudes from the pommel cap, positioned further from the terminal part of the underside of the lateral segment, where, in the case of some Type E and T1 swords, holes used for inserting a small pin to connect the upper guard with the pommel cap are located.
Fig. 35. A sword and a pommel cap of a possible Type T2:
a. Doles Vampenieši, grave 170 – VI 124:1758 (after Tomsons 2018: Fig. 25.1);
b. Salaspils Laukskola, grave 300 – VI 128:3405 (after Tomsons 2018: Fig. 25.4).
In the region of former East Prussia, we record only one sword from which the pommel and crossguard have been preserved, originating from the cemetery at Linkuhnen (Goßler – Jahn 2019: 185–186). To this find, we can add two further pommel caps of unknown provenance that were discovered within former East Prussia and are currently housed in the collections of the State Museums in Berlin – Prussian Cultural Heritage Foundation (SMB – Preußischer Kulturbesitz) under inventory numbers PM Pr 9165/8 and PM Pr 9119/1. According to the description of the objects in the online catalogue, these pommel caps correspond to Types T/T1 according to Kazakevičius. From the photograph published in the SMB portal catalogue, it can be seen that the surface of the pommel cap under number PM Pr 9165/8 is incised with vertical grooves, which in the past were likely filled with a non-ferrous coloured metal. The overall condition of the object is relatively poor; one lateral segment and the apex of the central segment have not survived. The specimen can tentatively be classified as an uncertain Type T2 based on the prominent lateral segment in the shape of an animal head, which is similar to that on the pommel cap of the Norwegian specimen from Momrak. The second find has been preserved in a somewhat better condition. Although it is unclear from the photograph on the SMB portal whether the pommel cap retains any fragmentary decoration, its morphological attributes are close to Type T2; consequently, we have chosen to designate this specimen as a possible Type T2.
According to the research team of Maczek, Môc, Dvorák, and Beronská, not a single Type T2 sword occurs within the territory of Lithuania (Maczek et al. 2024: 293). However, based on the aforementioned analysis of the sword from the site of Viešvilė in the Jurbarkas district, we infer that one specimen of Type T2 originates from Lithuania.
In the Old Rus’ environment, Anatoly N. Kirpichnikov recorded only four Type T2 swords (Kirpichnikov 1966a: Cat. Nos. 39–42). The researcher also assigned a find from Zaozerye in the south-eastern Lake Ladoga region to this type (1966a: Cat. No. 40), which, however, according to Sergey Kainov belongs rather to Type V (Shchedrina 2023: 94, Note 6; Viskupič 2025; Personal communication with Sergey Kainov). According to Sergey Kainov, a total of eight Type T2 sword finds are known from the territories of present-day Russia and Ukraine (Kainov 2012: 39). The research team of Maczek et al., based on morphological and decorative features, lists only four finds from Russia and three from Ukraine, classifying only the specimen from Dneprostroi as a certain type, while the remainder are considered uncertain Type T2 and/or interpreted as transitional or hybrid T2/V variants (Maczek et al. 2024: 293–294, Fig. 16, Cat. Nos. 6, 11, 13–17).
We believe that five specimens discovered in the territory of present-day Russia can be assigned to the type under study. As many as three swords originate from the Leningrad Oblast in the north-west of the country – Vakhrushevo, Leonovo, and Novinka (Kirpichnikov 1966a: Cat. No. 39; 1992: Cat. No. 1; Kochkurkina – Summanen 2021: 251–253; Shchedrina 2023: 94). The last-mentioned sword originates from a destroyed kurgan and was preserved with an incomplete blade, broken approximately in half, and with both structural components of the hilt, though without decoration (Shchedrina 2023: 94, Fig. 3.1). On the pommel and crossguard, incisions are discernible which in the past were filled with non-ferrous metal wire and are visible today in some places in the form of droplets. This indicates that the sword was likely part of a cremation ritual, which caused the coloured infill to melt (Shchedrina 2023: 94, Fig. 4.2). The incisions are executed predominantly vertically on both guards, whereas on the pommel cap they run in various directions (Shchedrina 2023: 94, Figs. 3.2–3, 4.1–2). Alexandra Shchedrina comparatively correlates the pommel of this sword with the specimen from Kjølstad Vestre, Norway (C37550/a) (Shchedrina 2023: 94, Note 5). In the close-up shots of the pommel, we can observe that the central segment is shaped like a triangle with a rounded apex, and the lateral segments are characterised by a high ridge and a prominently concave bridge, further underscoring their zoomorphic character.
Another find comes from the Republic of Karelia, specifically from the village of Shuya in the Prionezhsky District; this is a sword preserved almost completely, with the exception of a small part of the tip (Shchedrina 2021: 270–273, Fig. 1). The fourth find is a sword with a missing portion of the blade from the destroyed barrow in Gnezdovo, Smolensk Oblast, which A. N. Kirpichnikov originally classified under Type V (Kirpichnikov 1966a: Cat. No. 62; Kainov 2012: 39–41; 2019: 61–63).
Fig. 36. a1. A Petersen Type T2 sword from the Shuya site (after Shchedrina 2021: Fig. 1);
a2. close-up of the hilt of the sword from the Shuya site (after Shchedrina 2021: Fig. 1);
b. a Petersen Type T2 sword from the Novinka site, Leningrad Oblast, Russia (after Shchedrina 2023: Fig. 3.1);
b1. view of the pommel of the sword from the Novinka site (after Shchedrina 2023: Fig. 3.2);
b2. view of the crossguard of the sword from the Novinka site (after Shchedrina 2023: Fig. 3.3).
From the territory of Ukraine, we record three finds of Type T2 swords, two of which were discovered in the Dnieper River in the Zaporizhzhia and Kherson Oblasts (Takhtay 1935: 130–133; Chernyshev 1963: 221–223; Kirpichnikov 1966a: Cat. Nos. 37, 41; Komar 2014: 56, Cat. No. 4). The third specimen is a fragmentary sword originating from the Black Grave in Chernihiv, in the north of Ukraine (Kirpichnikov 1966a: Cat. No. 41; Androshchuk – Zotsenko 2012: 288).
In the region of Central Europe, we record only a single find of a Type T2 sword, which was discovered in the early 1970s in eastern Slovakia within the cadastral area of the then village of Krásna nad Hornádom, now part of the city of Košice (Košice – Krásna nad Hornádom) (Baxa 1974: 60–62, 109; Ruttkay 1975: Cat. No. 79A-1; Môc 2018: 221–233; Maczek 2024: 277).
Distribution map of Petersen Type T2 swords:
blue – certain Type T2; red – uncertain Type T2; yellow – swords of an uncertain Type T2 with an unknown find-spot.
Find Context and Dating
Out of the total of 41 swords analysed above, we can therefore assign 27 specimens to the certain Type T2 based on their morphological and decorative attributes, while another 11 specimens are regarded as uncertain types. Another three Latvian swords (Gramzdas Upītes, Kokneses kapulauks, Turaidas Pūteļi) were excluded from the catalogue due to the unsatisfactory conclusions of our observation.
The find context of both swords mentioned by Petersen (C1648, C17958) is essentially uncertain. The first of these, the sword from Utgården, is recorded on the Unimus portal as a grave find without a specified category. Similar to this description, Jan Petersen states that the sword was found with three rattles of a later type – C1649, C1650, and C1651 (Petersen 1919: 152). These are iron objects consisting of a main ring with several additional rings attached to it (Kåresen 2024: 5, Fig. 1.5). The rattles themselves constitute a group of intriguing objects whose exact significance has not been fully clarified, but they are assumed to have been related to horse equipment; specifically, their function is associated with use on sledges as part of a horse harness (Petersen 1951: 43; Kåresen 2024: 5). The majority of finds originate from the eastern part of Norway (Petersen 1951: 43) and date to the period between 750 and 1050 (Kåresen 2024: 5). Jan Petersen dedicated a short section to them in his spectacular work on Norwegian Viking Age swords (1919: 48–50, Figs. 46–50), where he bypassed their functional use and focused more on their typological and chronological development in relation to other objects. He paid more significant attention to them in his later work Vikingetidens Redskaper (1951: 42–56, Figs. 40–46). A detailed analysis was produced by Beate Krydsby Kåresen in her Master’s thesis, based on the methodology and theory of experimental ethno-archaeology (2024). The researcher distinguishes two groups depending on their structural parts (Kåresen 2024: Figs. 1.1–2, 1.5–6), typologically adhering to the classification by Rygh and Petersen while adding four new categories: BK1, BK2a, and BK2b (Rygh 1885: R 460–463; Petersen 1951: Figs. 40–46; Kåresen 2024: 36, Appendix 1). All three rattles originating from Utgården belong to Type R461, with the specimen under number C1649 being supplemented by a small bell (Petersen 1919: Fig. 50; Kåresen 2024: Fig. 8.9). The type can be defined as rattles with an attached hook, without a handle, and with the main ring in two parts, meaning these finds should belong to the second group – rattles without a handle (Kåresen 2024: Appendix 1, R461, Figs. 1.2, 6). Petersen classified them as a chronologically later type (Petersen 1919: 152). According to him, Type R461 was represented by 20 specimens, of which less than half came from grave finds; however, he notes that not a single one dates prior to the year 1000 (Petersen 1951: 54). In the description of the finds (Funnbeskrivelse) on the Unimus portal, it is mentioned that the objects were discovered in the ground on the property of the Utgården farm in the municipality of Seljord, Telemark county, with the addendum that the objects might have originated from the neighbouring Nordgaarden estate, where burial mounds were allegedly located.
The second sword was discovered in the summer of 1894 during roadworks in Øvrebø, near the municipality of Bø (also in Telemark county), at a depth of half a metre, together with three axe heads of Type Rygh 556—C17961, C17962, and C17963. The description for the first find states that the heads belong to a single type with varying shapes. The heads exhibit the closest similarities to Petersen Type K (personal communication with Tomáš Vlasatý). According to Petersen, this type varies in form, which complicates its typological and chronological development, but he nevertheless inclines towards a dating of its use during the 10th century (Petersen 1919: 44–45, Figs. 41–42). Other sources date these axes in a narrower scope from the second quarter of the 10th century to its end, and in a broader framework from the beginning of the 10th century to approximately the end of the second quarter of the 11th century (Hjardar – Vike 2016: 163). The remaining part of the finds consists of various iron objects (C17959/a–e, C17960). Based on the description from the Unimus portal, we cannot safely confirm that these objects originate from a single closed grave assemblage.
A possible grave find is also represented by the sword from Mårem Nordre, Tinn in Telemark county (C52806/1). According to the description available on the Unimus website, the sword, along with several other objects, was discovered in the 1990s during potato harvesting in a field within an area of approximately 4 × 4 m, but without obvious traces of a grave. Accompanying the sword were two axe heads of Petersen Types K (C52806/2) and G (C52806/3), a spearhead (C25806/4, which according to Unimus is similar to Petersen Type G), a knife, and an arrow-head (C52806/5; C52806/6). Axes of Petersen Type K (hereinafter referred to only as Type J.P.) date, as stated above, to the 10th century with a possible extension into the early 11th century. Conversely, axes of Type J.P. G represent a chronologically earlier variant, the dating of which is given between 825 and 925, and in a broader chronological framework even between 800 and 975 (Vlasatý 2016b; Hjardar – Vike 2016: 163). In the case of the spearhead, we are uncertain about the correctness of the cataloguing on the Unimus portal, where it is stated that this object is similar to Type J.P. G (Petersen 1919: Fig. 17). The preserved part of the socket, however, is relatively narrow and long; the shoulders of the blade flare widely, and the cutting edge then curves into the widest section, from where it tapers towards the tip. A ridge passes through the centre of the blade, giving the spearhead a lozenge-shaped cross-section. We have been unable to trace close analogies, which practically prevents establishing a chronological horizon for this object.
From the Momrak farm in Fyresdal, Telemark county (C28352/a), a sword originates from a male grave, discovered together with an axe head (C28352/b). Based on the description on the Unimus portal, this head should belong to Petersen Type L (Petersen 1919: Fig. 43), with the addendum, however, that the spurs are similar to those of Type G (Petersen 1919: Fig. 38). We believe that the head could also be similar to Petersen Type H. This variant dates between 900 and 975 (Vlasatý 2016b; Hjardar – Vike 2016: 163). Type L represents a chronologically later variant with a presumed start of use from the second half of the 10th century, extending into the 11th century (Petersen 1919: 45); other sources suggest the occurrence of axes of this type as early as the end of the first half of the 10th century, with possible continued use into the second half of the 12th century (Hjardar – Vike 2016: 163).
Fig. 37. A Petersen Type T2 sword and an L/H type axe head from Fyresdal – C28352 (adapted from unimus.no).
The sword discovered in Kjølstad Vestre, Sør-Odal, Innlandet county (C37550/a), as stated in the information on the Unimus portal, is said to originate from a Late Viking Age male grave. It is further noted that it was likely a cremation burial, and there are no indications of a burial mound. Among other items of weaponry, a spearhead of Petersen Type K (C37550/b), an axe head of Petersen Type H (C37550/c), two shield bosses of Type R565 (C37550/d; C37550/e), and arrow-heads of various shapes (C37550/f–i) were present. In addition, equestrian equipment, a sledge, and other objects were found alongside the aforementioned items (Vlasatý 2026). According to the Norwegian researcher, spearheads of Type J.P. K appear in combination with other items of weaponry that can be dated to the Late Viking Age (Petersen 1919: 33). In the work of Bergljot Solberg focusing on Merovingian and Viking Age spearheads (1984) – where the author analysed a significant quantity of these objects divided into twelve basic groups (designated by Roman numerals) with several sub-groups – she states that her subtype VII.2B (J.P. K) appears as early as the first half of the 10th century, with its use continuing until the end of the 10th century (Solberg 1984: 86, 95–96; Hjardar – Vike 2016: 175). Based on Petersen’s research, shield bosses of Type R565 can be dated similarly to Type R563, i.e., approximately from the second half of the 10th century with an extension into the medieval period (Petersen 1919: 47). Other sources assume the occurrence of this type of boss around the fourth quarter of the 10th century, remaining in use after the year 1000 (Pedersen 2014: Fig. 5.11; Hjardar – Vike 2016: 185).
Fig. 38. A Petersen Type T2 sword alongside shield bosses, a spearhead, and an axe head from Kjølstad Vestre – C37550/a (adapted from unimus.no).
The sword from Sømhovd, Sømna, Nordland originates, according to the description on the Unimus portal, from a chamber grave situated in the middle of a large burial mound, likely 6 m high. Of the bones, an unburnt human cranium was discovered. In addition to the fragmentary sword, iron fragments were recovered from which pieces of a shield boss similar to Type R565, numerous rivets, a large arrow-head similar to Type R539, and a fragment of a circular part of a bit were identified (C13816/1–4). A spindle whorl made of green glass (C13817) was also present in the grave, suggesting that two individuals of different sexes may have been buried in the mound.
Jan Petersen dated Norwegian Type T swords between the second half of the 10th and the early 11th century (Petersen 1919: 152, 183). For both specimens from Telemark (C1648, C17958), he also adds that their dating is not unequivocal; in particular, the sword from Øvrebø along with the axe heads is impossible to date precisely, as these represent mixed objects (Petersen 1919: 152). The sword from Seljord could only be tentatively assigned to approximately the first half of the 11th century in connection with the rattles of the later type. The find of the sword from Mårem Nordre also appears uncertain, as we cannot confirm whether it genuinely originates from a grave. Due to the ambiguity of the find circumstances, this sword is difficult to date. In our opinion, based on the axe head, the grave from Fyresdal cannot be dated more precisely than to the second half of the 10th century or the first half of the 11th century, which in this case would also be supported by the inscription on both sides of the blade (+INGERIIIFECIT/CONSNVIIINS). This text should belong to the Ingelrii group of inscriptions, which likely appear in the second half of the 10th century, with their use extending into the first half of the 11th century, or possibly even further, to the period of 1150–1200 (Moilanen 2015: 140). With this sword, however, we cannot rule out the possibility that the structural parts of the hilt were mounted onto a newer blade. Regarding the dating of the grave from Kjølstad Vestre, the assemblage of objects representing weaponry indicates that the grave likely dates to the close of the 10th century. The occurrence of the Type J.P. K (Sol. VII.2B) spearhead precludes the possibility that it dates to the period after the year 1000. Hanne Lovise Aannestad also inclines towards the same dating (Aannestad 2018: 157). The chamber grave from the Sømhovd mound can only be tentatively dated, on the basis of the find of the Type R565 shield boss (C13816/1), to the period of the second half of the 10th to the early 11th century, because the object is present only in a few small fragments from which it is difficult to evaluate its original shape.
From the data presented above, it can therefore be stated that the assignment of Norwegian Type T2 swords according to Petersen to a specific chronological horizon remains unsatisfactory despite several grave finds, and we were able to determine the approximate formation of the grave assemblage for only one specimen. At the same time, we believe that this type does not appear in Norway prior to the course of the second half of the 10th century, and some specimens could have been deposited in graves at least as late as the beginning of the 11th century.
In Finland, we register only two Petersen Type T2 swords discovered in graves at the cemetery in Vanhakartano—KM 8602:130 and 141 (Cleve 1978: 59–60; Moilanen 2015: 387; 2018: 125). The first sword originates from inhumation grave C I, investigated by Alfred Hackman in 1925 (Moilanen 2015: 454). Accompanying the sword in the grave were other objects – an axe head (the type of which we could not verify), two fragments of unspecified silver coins, five iron weights coated with bronze, a bronze object on which fragments of woollen fabric and (likely) leather were preserved, a bronze spiral ring that had become attached to the sword blade due to corrosion, a fragment of a scythe, iron shears, peas, and other organic material (Cleve 1978: 59). Nils Cleve dates the grave based on the coins to the period between 1025 and 1150 (Cleve 1978: 196–197). The second sword comes from inhumation grave C K, where remnants of wood and likely also bark from the coffin were preserved (Cleve 1978: 59). The sword was present in the grave together with two spearheads, a bronze horseshoe-shaped brooch, fragments of a bronze buckle and a bronze belt chape, a bronze circular belt distributor, a knife with a bronze scabbard fragment featuring stamped circular decoration, four fragments of unspecified silver coins, and several other iron and bronze objects; remnants of organic material were also present (Cleve 1978: 60). The dating of the grave according to N. Cleve falls, based on the coins, within the period of 975–1025 (Cleve 1978: 195–196).
The spearhead with number KM 8602:142 (Cleve 1978: 60, Cat. No. 312, Fig. 20.312) is, according to N. Cleve, likely a Petersen Type G, though it exhibits similarities to Type J.P. M, or alternatively to Type J.P. K. Type J.P. G (Sol. IX.1A) is characterised by a short, thick socket with a wider expansion at the lower end; the transition from the socket to the blade is short, and the blade itself is wide at the shoulders, from where it tapers evenly towards the tip (Petersen 1919: 29, Fig. 18; Solberg 1984: 120, Fig. 16). According to Petersen, these are spearheads occurring from the second half of the 10th century until the 11th century, a view with which Bergljot Solberg concurs (Petersen 1919: 29; Solberg 1984: 124). In Finland, roughly 120 specimens of this type have been recorded, 54 of which feature sockets decorated with silver in the zoomorphic Urnes style, indicating a dating as late as the second half of the 11th century (Lehtosalo-Hilander 1982: 32–34; Solberg 1984: 124). The shape of the blade of the Finnish specimen is relatively narrow and long, making it appear closer to Type J.P. K; the socket appears more robust at the base than is common for this type (Petersen 1919: Fig. 21), but it is longer and the transition to the blade is smooth, passing into a concave curve. Type K is represented in Finland by 60 finds (Lehtosalo-Hilander 1982: 32). The second spearhead – KM 8602:143 (Cleve 1978: 60, Cat. No. 313, Fig. 20.313) – should, according to him, be closest to Type K. Morphologically, this specimen is also close to Type J.P. I (Sol. VII.2A), which is characterised by a long, narrow socket with a widened opening for the shaft at the lower end and a long, narrow blade (Petersen 1919: 31, Fig. 20). However, all specimens of this type are supplemented by small bronze rivets positioned along the lateral sides of the socket (Solberg 1984: 85). In the case of the Finnish specimen, neither rivets nor holes for them were recorded. Type I occurs predominantly in the first half of the 10th century, with some Norwegian specimens dating to the 10th century in general (Petersen 1919: 31; Solberg 1984). Some later variants with a shorter blade date as late as the third quarter of the 10th century, such as two finds from Valsgärde (boat chamber graves 12 and 15) in Uppland, Sweden (Sénby-Posse 2021: 65–66, 91–92; Vlasatý 2021; Viskupič 2025) or one from Gjermundbu, Buskerud county, Norway – C27317 (Solberg 1984: 94–95). Another problem in this case is the absence of Type I in Finland and the Baltic region (Solberg 1984: 105–106).
Fig. 39. a–c. Selected objects from grave C I at the Vanhakartano cemetery (after Cleve 1978: Pl. 19.298, 301, 302);
d–h. selected objects from grave C K, Vanhakartano (after Cleve 1978: Pl. 20.311–13, Pl. 21.314–17).
The remaining two finds are stray finds, of which the sword under number KM 3699:3 was discovered with charcoal. Mikko Moilanen assumes that it may originate from a cremation cemetery (Moilanen 2015: 369). The second sword (KM 31017:1) is a detector find from the late 1990s, which may likewise originate from a cremation cemetery (Moilanen 2015: 418). The dating of Type T swords within the territory of Finland is placed in the period from the second half of the 10th century until the first quarter of the 11th century (Moilanen 2015: 253). Both grave finds from the Vanhakartano cemetery were dated according to coin finds, which, however, are not further specified. This complicates our verification of the graves’ formation within a specific chronological horizon.
Mati Mandel dates Type T2 swords from Estonia to the 10th century based on Old Rus’ material, listing only one stray find of this type from the Kalmumäe cemetery, Upa village on the island of Saaremaa, discovered together with spearheads, which allegedly occurred in 1961 (Mandel 1991: 117). Given the unclear find context of both Estonian specimens from the island of Saaremaa, we can date these objects only within a broader chronological framework between the second half of the 10th and the first half of the 11th century.
According to Artūrs Tomsons, Latvian sword finds of the group under study can be dated from the 10th to the first half of the 11th century (Tomsons 2018: 132, Fig. 101.2). He links their occurrence in this territory with international trade of the 10th to 11th centuries, pointing out that the swords originate from coastal areas and from sites located along the largest inland waterways (Tomsons 2018: 64–65). We managed to verify the find context for only four of them. The detector find of a sword from the lower reaches of the Dvina River may, according to A. Tomsons, originate from a cremation grave in areas inhabited by the Livonians, or alternatively from a site where Curonian tribes occurred (Tomsons 2018: 62). It was reportedly discovered together with a scabbard chape featuring a motif in the Scandinavian Borre art style, a lanceolate spearhead, and belt components, from which a buckle, five mounts, and an oblong strap-end with a curved terminal were present (Tomsons 2018: 62). All objects were published by the finder in May 2008 on the antik-war.lv online forum, meaning that their discovery must have occurred either during that same year or slightly earlier. It is clear from the photographs that at the time of discovery, the sword had a complete blade present in three pieces and with a noticeable bend, which may indicate that it was ritually deformed, as pointed out by A. Tomsons (Tomsons 2018: 62). The openwork scabbard chape, likely cast from a copper alloy, bears the motif of a stylised bird viewed from above with outstretched wings and tail, intertwined with lines of bands and tendrils. The upper part is shaped into a zoomorphic form with the indication of eyes in the form of two circles and a narrow, long, rounded apex resembling a beak or snout. The closest analogies originate predominantly from Scandinavia, especially from Sweden, followed by Norway, Island, and Denmark, including Schleswig-Holstein (Paulsen 1953: 18, 20, Figs. 5–7). Another two finds are also known from Latvia (Tomsons 2018: 142–143, Fig. 109.9–10). Peter Paulsen classified this type of chape under the type of “chapes with a Germanic bird motif” (Ortbänder mit germanischem Vogelmotiv) and within the group of so-called “chapes of the Scandinavian group” (Eine Skandinavische Gruppe) (Viskupič 2023). He saw their origin precisely within the Old Norse environment, while also pointing to their imitations or modified forms emerging outside this region, particularly in East Prussia around the year 1000 (Paulsen 1953: 18, 20, Figs. 8–10). According to P. Paulsen, the bird pattern is modelled in the Borre style, and based on grave finds, he dates the Scandinavian forms to the second half of the 10th century (Paulsen 1953: 20). Fedir Androshchuk disagrees with the rendering of the motif, suggesting that the pattern represents a bird-beast with clearly indicated paws and outstretched wings, which he classifies under his Type 3c (Androshchuk 2014: 116, Fig. 71c). At the same time, he believes that the art style, given the double contours and the pelleted representation of the wings, is related rather to the Jelling and Mammen styles (Androshchuk 2014: 120). He views this type as later and dates it between the years 950 and 1000 (Androshchuk 2014: 120–121).
The spearhead has a short blade with a lozenge-shaped cross-section. From the tip, it widens slightly towards the middle into a smooth oval and further towards the socket it narrows at the neck. The socket widens downwards. The shaft hole is likely circular. Within Petersen’s typology, the lanceolate spearhead is closest to Type J.P. E (Petersen 1919: 26–28, Figs. 12–13), or alternatively Type VI.4 according to Bergljot Solberg, which the researcher divided into three subtypes based on cross-section (Solberg 1984: 58, Fig. 11). In her typology, this spearhead may belong to subtype VI.4C. Anatoly N. Kirpichnikov assigned this form to Type I (Kirpichnikov 1966b: 9, 12, Fig. 1). It is one of the most widespread types of spearheads across Viking Age Europe (Kainov 2019a: 121–122). In Latvia, such spearheads have been found in all parts of the country, with Māris Atgāzis classifying them under Type A, divided into four subtypes according to the length and width of the blade and socket (Atgāzis 2019: 46–48, Figs. 14–15). The dating of this type of spearhead is given as the period between 800 and 950 in the case of Norway (Solberg 1984: 68). In the Old Rus’ environment, they occur from the end of the 9. to the first half of the 11th century (Kirpichnikov 1966b: 9; Kainov 2019a: 122). According to M. Atgāzis, the Latvian specimens are dated between the 9th and the early 11th century, with their greatest popularity recorded between the second half of the 10th and the first half of the 11th century (Atgāzis 2019: 46, 48). Given the nature of the find circumstances, the objects can only be tentatively dated to the period of the second half of the 10th to the early 11th century.
Another two detector finds, pommel caps, were discovered alongside several objects. The pommel cap originating from an unknown location somewhere on the lower reaches of the Venta River was found together with a silver twisted horseshoe brooch with zoomorphic terminals and a lozenge-shaped object, the outer sides of which are modelled in the form of pelleted decoration. This is likely a pendant or amulet (Tomsons 2018: Fig. 30.1). On the basis of these objects, A. Tomsons dates the pommel cap to the 10th to 11th centuries (Tomsons 2018: 64). The second pommel cap, likely from the region of Livonia, was reportedly found together with a bronze triangle-headed pin in the Gauja River basin (Tomsons 2018: 64, Fig. 30.2).
From the Salaspils Laukskola cemetery comes the find of a pommel cap, which is designated in the literature as Type T1 but exhibits morphological traits similar to Type T2 (Tomsons 2018: 62, Fig. 25.4). The object was found in cremation grave 300, and besides a heap of burnt bones, a quantity of fragmentary iron and bronze objects occurred there (Zariņa 2006: 92, Fig. 168). A. Zariņa dates some of the grave finds from this cemetery from the second half of the 11th to the first half of the 12th century (Tomsons 2018: 64–65). Among the objects from this grave is a fragment of a single-edged blade, which was identified as a sword blade (Zariņa 2006: 92). Whether it has any connection with the pommel cap, however, is unknown.
.Fig. 40. Detector finds of swords/sword fragments and other objects from various parts of Latvia:
a. the lower reaches of the Dvina River (adapted from antik-war.lv);
b. the lower reaches of the Venta River, Curonia (after Tomsons 2018: Fig. 30.1);
c. the Gauja River basin, Livonia (after Tomsons 2018: Fig. 30.2).
The only East Prussian specimen of the studied sword group with atypical decoration originates from the cemetery at Linkuhnen, now Rzhevskoye in the Kaliningrad Oblast (Russian Federation). The sword, which at the time of discovery also had its blade present in several pieces, comes from cremation grave 53, investigated in 1929 (Goßler – Jahn 2019: 185). Within the accompanying inventory of this grave, we can mention a sword pommel of Petersen Type X, an iron spearhead with a bent blade, and two pairs of circular bits with a twisted mouthpiece, one of the bits retaining bridle strap mounts on its rings. Furthermore, three iron stirrups, a bronze pin with a polygonal head and a similarly shaped central section, a fragment of a twisted bronze neck-ring, a fragment of a neck-ring made of bronze sheet metal, a fragment of a bronze bracelet, a hammer-shaped belt chape, and an ornate bronze mount were present (Goßler – Jahn 2019: 361, Figs. 41.1–5, 42.1–2, 43.1). Petersen Type X swords represent relatively simple forms consisting of straight, long crossguards and a single-part semi-circular pommel, which the Norwegian researcher divided into two groups based on the shape and proportional attributes of the hilt’s structural parts (Petersen 1919: 159, Figs. 124–129). Swords of this type were distributed across a wide geographical area, from Western Europe through Scandinavia, Central, and Eastern Europe to South-Eastern Europe (Viskupič 2026b). Approximately eighteen finds, including grave 53/1929, originate from Linkuhnen (Goßler – Jahn 2019: 196, Tab. 20). Jan Petersen dated the Norwegian specimens between the first half of the 10th century and the 11th century (Petersen 1919: 182). Great Moravian swords of Type J.P. X date from the second half of the 9th to the early 10th century (Hošek – Košta – Žákovský 2021: 305). Their use in the Baltic region dates between the 10th and 11th centuries, and at Linkuhnen, according to Norbert Goßler and Christoph Jahn, as late as the end of the 10th century (Goßler – Jahn 2019: 196, Tab. 20). Vytautas Kazakevičius extends the dating of Type J.P. X for the Baltic region up to the 12th century (Kazakevičius 1996: 138). All three stirrups belong to Goßler’s Type BI1a. Their origin is placed within the nomadic equestrian environment of the Eurasian steppes, with dating from the 9th to the 11th or potentially 12th century (Goßler 2013: 136). They are known from several sites in East Prussia, including Linkuhnen, and the find context of these objects indicates a broader dating into the 10th to 12th centuries (Goßler 2013: 136, Fig. 4). The grave assemblage 53/1929 itself, in light of the Type T sword, is dated by Goßler and Jahn to the period around the year 1000, or more precisely to the end of the 10th to the early 11th century (Goßler – Jahn 2019: 198, 361).
Fig. 41. Assemblage of objects originating from grave 53/1929 at Linkuhnen (after Goßler – Jahn 2019: Pl. 43.1).
In the Old Rus’ environment, eight Petersen Type T2 swords are known, six of which originate from graves, the find context of two is associated with bodies of water, and one is a stray discovery (see Appendix). Anatoly N. Kirpichnikov dates this type to the 10th century (Kirpichnikov 1966a: 30).
The sole Gnezdovo sword of Type J.P. T2 featuring stepped decoration originates from a destroyed barrow; it survived with its complete hilt and a fragment of the blade (Kainov 2012: 39–41; 2019a: 59–63). It was discovered by labourers during the demolition of the barrow, with the weapon located at its base, stuck by its blade into the ground (Sizov 1902: 28; Kainov 2012: 39; 2019a: 61). Given the fact that the other objects from the destroyed barrow have not survived, dating its archaeological context is impossible (Kainov 2012: 41; 2019a: 63).
Another sword originates from a destroyed barrow near the village of Leonovo, situated in the Leningrad Oblast in the south-eastern Lake Ladoga region (Kochkurkina – Summanen 2021: 20). The barrow was part of a group of three or five mounds situated between the right bank of the Pasha River and the left bank of a stream flowing into the river, above the village of Leonovo in a pine forest (Kochkurkina 1989: 172–173; Kochkurkina – Summanen 2021: 20). From this barrow comes a sword with its blade broken into two pieces, retaining its crossguard and tang but lacking its pommel, alongside an iron axe head and a fragment of an iron spearhead (Kochkurkina – Summanen 2021: Fig. 21.1–4). Due to the missing pommel, the sword was designated as a possible Type T2 or V, or alternatively a transitional type (Kochkurkina 1989: 173; Kirpichnikov 1992: Cat. No. 1; Košta 2021: Tab. 10, Cat. No. 2265; Maczek et al. 2024: 293). Following analysis of the crossguard decoration, however, it is certain that the sword belongs to Type J.P. T2 (Kochkurkina – Summanen 2024: 252, Figs. 1.1, 3, 2.1, 2, 3) (Fig. 14c1–2). According to S. I. Kochkurkina, the axe head belongs to Kirpichnikov Type V (Kochkurkina 1989: 173; Kochkurkina – Summanen 2021: Fig. 21.4). Kirpichnikov states that it is possible to distinguish two subtypes – battle-axes (Kirpichnikov 1966b: 37, Figs. V.3, XIV.1–4, XVII.5, XXIII.5) and working axes (Kirpichnikov 1966b: 37, Figs. XVI.3, 6, XVII.9). From the Old Rus’ environment, this type is most numerous in the northern section, where it dominates mainly in the barrows of the south-eastern Lake Ladoga region (Kirpichnikov 1966b: 38). He associates their origin with Scandinavia and, due to their significant numbers in Finland, terms them “Finnish-Russian” (Kirpichnikov 1966b: 38). In Petersen’s typology, they are close to variant C, which appears as early as the Roman period (Petersen 1919: 39, Fig. 32; Kotowicz 2018: 64), but as Piotr Kotowicz correctly notes, the specimens from the barrows of the south-eastern Lake Ladoga region, in comparison with their Scandinavian counterparts, are characterised by a wide beard, and can therefore be considered more distant analogies (Kotowicz 2018: 64). Similarly, Sergey Kainov points out that the Old Rus’ specimens are a later evolutionary form of Petersen Type C (Kainov 2019a: 149–150). According to Kirpichnikov, this type of axe appears in the 10th century and their use lasts until the first half of the 12th century, but the earliest specimens appear in Rus’ at the end of the 8th to the early 9th century (Kirpichnikov 1966b: 37; Kainov 2019a: 149–150). He further adds that the 10th-century finds are associated primarily with Finnish sites in the north-east (Kirpichnikov 1966b: 37). The fragmentary spearhead, of which only the blade has survived, appears similar to the long lanceolate forms corresponding to Petersen Type E, or alternatively Type VI.4 according to B. Solberg (Kirpichnikov Type I); however, from the state in which it was preserved, this cannot be unequivocally determined (Kochkurkina – Summanen 2021: 256, Fig. 8.1). The dating of the barrow based on the sword and axe can thus be estimated between the second half of the 10th and the first half of the 11th century.
Fig. 42. Assemblage of objects from the destroyed barrow at Leonovo.
After Kochkurkina – Summanen 2021: Fig. 21.
From the Leningrad Oblast also comes a grave find of a sword discovered in a barrow near the village of Vakhrushevo. According to N. E. Brandenburg, seven closely situated mounds stood above the village of Vakhrushevo, with another two positioned aside on the right bank of the Pasha River, two of the barrows being investigated by him (Kochkurkina 1989: 178). Barrow CXVI contained the remains of burials in two layers – the upper layer contained the cremation burial of a man and a woman (together with a Type J.P. V sword), the remains of a cremated horse, accumulated bones without objects, the inhumation remains of an unburnt body, and three crania; the lower layer (Complex 8) contained the cremation burial of a male with accompanying grave goods (Brandenburg 1895: 129–130; Kochkurkina 1989: 178–179). Among the objects of the second layer was a Petersen Type T2 sword with a damaged blade (Kirpichnikov 1966a: Cat. No. 39, Fig. I.5). The remainder of the grave inventory consisted of: two axe heads, one of which was identified as Type II according to Kirpichnikov, a spearhead (of an unspecified type), three arrow-heads, a knife, and an iron handle for a whip with one end shaped as a lozenge with an aperture through which a ring passes, on which two loop-shaped pendants are suspended with their volute-shaped ends turned towards each other, and between them is a flat pair of sheet-metal tweezers secured with rivets. Furthermore, a fragment of a twisted silver bracelet, two padlocks and a key, a fragment of a large bronze horseshoe-shaped brooch, two spherical weights, fragments of a balance pan, a whetstone, and a fire steel with a zoomorphically shaped handle were present there (Brandenburg 1895: 129–130, Figs. III.23, VIII.8, IX.5, XI.4; Kochkurkina 1989: 179–180). Besides the male bones, the cremation burial also contained the bones of a horse, a bear, including nine bear claws, and the bones of a smaller animal (Brandenburg 1895: 129; Kochkurkina 1989: 179). Axe heads of Kirpichnikov Type II are characterised by a trapezoidal blade and an elongated hammer-like butt (Kirpichnikov 1966b: 35, Fig. 6). According to Kirpichnikov, the first specimens appear in the 9th century in Bashkiria and the Caucasus, where he places their origin; besides this, similar forms occur in smaller numbers also in the Carpathian Basin, Poland, former East Prussia, or Sweden (Kirpichnikov 1966b: 35; Kainov 2019a: 145). Sergey Kainov considers the specimens from the Old Rus’ environment to be possible local products given their significant occurrence within this geographical space, as well as their morphological homogeneity and formal divergence from the Bashkir and Caucasian axes, which are characterised by massiveness and slightly different proportions (Kainov 2019a: 145). A. N. Kirpichnikov dates the Old Rus’ specimens between the 10th and early 11th century (Kirpichnikov 1966b: 35). The dating of both male burials is placed within the 10th century, with the ethnicity of the individual from the grave containing the Type V sword being considered Scandinavian (Kochkurkina 1989: 180).
Fig. 43. Objects from barrow CXVI – lower layer, Vakhrushevo:
a. drawing of a Petersen Type T2 sword (after Brandenburg 1895: Pl. XI.4);
b1–2. sword and hilt close-up of a Petersen Type T2 (after Kirpichnikov 1966a: Pl. I.5, VII.3 – correct caption Pl. VIII.3);
c. iron whip handle (after Brandenburg 1895: Pl. IX.5);
d. fire steel with a zoomorphically shaped handle (after Brandenburg 1895: Pl. III.23);
e. iron(?) object – key(?) (after Brandenburg 1895: Pl. VIII.8).
From the south-eastern Lake Ladoga region in the Leningrad Oblast, we register an assemblage of objects originating from destroyed barrows near the village of Novinka on the Kapsa River, collected in 1957 by I. P. Krupeychenko. Among these were a Petersen Type T2 sword, designated in the work of S. I. Kochkurkina as Type J.P. T1, alongside a Type J.P. E spearhead (Kirpichnikov Type I), an axe head, two knives, and an arrow-head (Kochkurkina 1989: 84; Shchedrina 2023: 93, 94, Figs. 3–4, 5.1, 13.2). Given that the sword originates from a destroyed barrow, dating the archaeological context is impossible (Shchedrina 2023: 94).
The only grave find of the sword type under study from the territory of Ukraine is represented by a fragmentary sword discovered in the so-called Black Grave in Ukrainian Chernihiv. The barrow constitutes an important funerary monument in the history of early medieval Rus’, investigated between 1872 and 1873 by Dmitry Y. Samokvasov (Samokvasov 1916: 4–36; Kainov 2019b: 125; Kovalenko et al. 2020: 56; Murasheva et al. 2021: 1–2). The mound contained what was likely the cremation burial of two males of different ages and one female (Samokvasov 1916: 31; Rybakov 1949: 27; Murasheva et al. 2021: 2). The grave goods consisted of a vast quantity of objects, among which can be listed: two swords; two sabres; two helmets; two coats of mail; two sets of stirrups; several spearheads with inlaid sockets; two drinking horns from an aurochs decorated with silver mounts; an anthropomorphic figurine identified as a representation of the Old Norse god Thor (Þórr); two golden Byzantine coins, one of which survived fragmentarily and was minted between 945 and 959 during the reign of Constantine VII and Romanos II; two scepters, and female jewellery (Kainov 2019b: 126; Kovalenko et al. 2020: 57; Murasheva et al. 2021: 2–3, 9). Due to the poor state of preservation of the objects, a significant portion of which consists of a concretion of melted artefacts, we have chosen to focus briefly only on those most relevant for the purposes of this article. This clump was located in the northern part of the barrow’s hearth site, accumulated after the cremation pyre had burnt down; during cleaning, it broke into several parts and was originally composed of two swords, a sabre, two spearheads, an arrow, two knives, fragments of mail, stirrups, and a spur (Samokvasov 1916: 20–22, Fig. 23). Other objects – two helmets, two pieces of mail armour, two Byzantine gold coins, two horns, the anthropomorphic figurine, and a vessel with burnt bones – were removed from the pyre after cremation and placed on the upper section of the mound, built on top of the cremation site, which was subsequently covered over (Samokvasov 1916: 20–22; Kainov 2019b: 125–126; Kovalenko et al. 2020: 56; Murasheva et al. 2021: 3).
Fig. 44. Section of the grave inventory from the Black Grave in Chernihiv, Chernihiv Oblast, Ukraine. The fragment of the Petersen type sword is located next to the anthropomorphic figurine, beneath the front rhyton (after Murasheva et al. 2021: Fig. 2).
One of the swords was assigned to Type J.P. T2 (Kirpichnikov 1966a: Cat. No. 41; Androshchuk – Zotsenko 2012: 288), with its length allegedly reaching only somewhere around 80 cm, while the length of the second represented approximately 102 cm. This led B. A. Rybakov to the conclusion that one of the males buried in the mound was of a younger age and the second was an adult individual (Rybakov 1949: 39; Kainov 2019b: 127–128). The second sword was typologically assigned by A. N. Kirpichnikov to the special Type Z (Kirpichnikov 1966a: 34–35, Cat. No. 81). A recent revision, however, demonstrated that this sword has its closest analogies in the Western European environment, specifically in England, within the Wallingford group according to Vera Evison, which is an evolutionary stage of Petersen Type L swords dating to the late 10th to 11th centuries; Sergey Kainov presumes that this sword was most likely manufactured precisely on the territory of England (Kainov 2019b: 133, 137). The blade tang is covered by a hilt sleeve manufactured from almost pure copper, with its surface decorated with an interlaced ornament in two superimposed rows (Kainov 2019b: 130, Figs. 2, 4.2, 5.1). The closest analogies are known from the island of Saaremaa in Estonia, where a Petersen Type V sword features a sleeve with the same pattern manufactured from a copper alloy (Jets 2013: 47, Fig. 22; Kainov 2019b: 135), and the second analogy is a fragment of such a sleeve from the circular fortress of Trelleborg in Denmark, also made of copper alloy, though the surface is covered with silver sheet metal into which the ornament was incised (Kainov 2019b: 136, Fig. 10.2). Sergey Kainov presumes that the sleeve was attached later, onto the already finished sword (Kainov 2019b: 137).
Fig. 45. Fragment of the sword of the “adult individual” from the Black Grave:
a. current condition (after Kainov 2019b: Fig. 3.1);
b. drawn reconstruction (after Kainov 2019b: Fig. 5.1–6).
Of the three mentioned spearheads, one was subjected to research using synchrotron and neutron imaging methods, by means of which it was discovered that for exhibition purposes, the investigated spearhead had been glued together in the past from two or three distinct specimens (Kovalenko et al. 2020: 58–59, 63–64). The results of the analysis indicate that the blade portion of the head along with a small part of the socket formed a separate specimen, with cross symbols made of non-ferrous metal detected on both sides of two blade fragments (Kovalenko et al. 2020: 65, Fig. 3а–б, 11а–б). Two parts of the socket originated from the second specimen, which, as established by the research, was decorated with a geometric pattern filled with silver and a copper alloy (Kovalenko et al. 2020: 65). The pattern consists of vertically arranged, alternating triangles and lozenges, between which a motif of elongated stepped lozenges and triangles is integrated, the triangles being formed of twisted rods of silver and copper alloy (Kovalenko et al. 2020: 65–66, Fig. 11в, 13а–г). Furthermore, circular holes in two superimposed rows on both opposite sides, located within the lozenges, were revealed on the socket (Kovalenko et al. 2020: 65–66, Fig. 13в). There is thus no doubt that the socket was part of a spearhead belonging to Petersen Type I, or alternatively Type VII.2A according to Bergljot Solberg (Kovalenko et al. 2020: 66; Petersen 1919: Fig. 20; Solberg 1984: 84–85, Fig. 13). The dating for the Norwegian finds is given as the first half of the 10th century, but their use continued after the year 950 (Petersen 1919: 31; Solberg 1984: 94–95). The closest analogies of Type J.P. I spearheads with stepped decoration can be found on Gotland, which Lena Thunmark-Nylén classifies under her Type 2 (comprising Petersen Types D2, F, G, H, I, K, M), with two finds matching the socket from the Black Grave – Gothem (SHM 8479) and Alva (SHM 23473) (Thunmark-Nylén 1998: Pls. 243.3a–b, 244.1a–c; 2006: 302–304).
Fig. 46. a. Representation of the socket of the “second” spearhead from the Black Grave in Chernihiv (after Kovalenko et al. 2020: Fig. 3b–c);
b. X-ray radiograph of a smaller section of the socket of the “second” spearhead (after Kovalenko et al. 2020: Fig. 11b);
c. X-ray radiograph of a larger section of the socket of the “second” spearhead (after Kovalenko et al. 2020: Fig. 11c);
d. spearhead with a detail of the decoration from Alva, Gotland, Sweden – SHM 23473 (after Thunmark-Nylén 1998: Pl. 244.1a–c).
The grand composition of the grave goods indicates a high social status for the male individuals, who were buried in accordance with pre-Christian traditions at a time when this method of burial was already in decline. Raising the cluster of objects onto the original barrow mound clearly had a symbolic character, and since the clump exhibited an absence of female objects and the grave inventory of the buried woman itself was relatively poor, it is inferred that her social status may have been much lower (Kainov 2019b: 127). The two male individuals likely represented a father and son, which could explain the paired deposition of weapons and armour as well as the varying sizes of the swords, helmets, and drinking horns (Kainov 2019b: 127; Kovalenko et al. 2020: 56–57). The formation of the barrow was traditionally dated on the basis of a Byzantine coin to the 960s; however, radiocarbon analysis of samples from the mound determined its formation somewhat later, to the period of 980–1025 (Murasheva et al. 2021: 2).
The remaining three swords from the territories of present-day Russia and Ukraine represent stray finds. The almost completely preserved sword of the studied type from the site of Shuya in the Prionezhsky District was discovered at a depth of 40 cm below the ground surface, within a layer saturated with charcoal and other domestic waste, leading to the inference that the sword was in a secondary deposition context here (Shchedrina 2021: 270). According to Alexandra Shchedrina, a destroyed cemetery is located near the site of discovery, but it is uncertain whether the sword could have been related to it in any way (Shchedrina 2021: 270).
In the case of the sword from the Dnieper, we know that it is a river find which occurred in 1928 at the bottom of the Dnieper River in foundation trenches during the construction of the Dnieper Hydroelectric Station – Dnieproges/Dneprostroi (Ravdonikas 1933: 598, 614; Chernyshev 1963; Komar 2014: 61). The sword was found in an intact state, with the exception of a small piece of the tip, together with three other swords of Types R, S, T1 and a double-edged sword blade lacking a hilt. Some researchers associated these swords with the return of Prince Sviatoslav to Kyiv from his campaign against Bulgaria and Byzantium, when he and his retinue were ambushed by the Pechenegs; alternatively, they considered the swords to be trade merchandise from Scandinavian merchant vessels that sank as a result of an armed conflict, or these weapons were viewed as a ransom (Komar 2014: 47–48). A ritual motif of offering weapons into water is also taken into account, documented by numerous archaeological finds in Northern Europe, which, in conjunction with the accounts of Constantine Porphyrogenitus regarding the sacrificial site of the Rus’ on the island of Khortytsia, made it possible to assign the swords from Dneprostroi to the traces of ritual practice (Androshchuk 2001: 130–132; Duczko 2004: 251; Komar 2014: 48–49). Similarly, a river find is also the sword discovered between Kherson and Oleshky, which occurred in January 1932 by fishermen who were casting nets under the ice; the sword was subsequently handed over to the Kherson Museum (Takhtay 1935: 130).
The “Type T2” Sword from Awajki
The sword is housed under inventory number PM Pr 347 in the collections of the State Museums in Berlin – Prussian Cultural Heritage Foundation (SMB – Preußischer Kulturbesitz). It was discovered in the 1840s during construction work on a canal along the shore of Lake Pinnau in former East Prussia (Petri 2017: 150), currently Lake Piniewo in the Warmian-Masurian Voivodeship, Poland. The find-spot is cited as Awecken (now the village of Awajki, in the western part of the Warmian-Masurian Voivodeship); in addition, the name Draulitten (now Drulity, near Awajki) is also mentioned, or alternatively, it is recorded simply as an unknown locality within former East Prussia (Petri 2017: 150). At the time of its discovery, the sword was completely intact, although an approximately 15 cm long section of the blade tip later broke off and was subsequently reattached using two small metal plates (Petri 2017: 151, Pl. 5b). The pommel was composed of two components – a three-lobed pommel cap and an upper guard – which were reportedly secured to one another by a curved rivet (Petri 2017: 151–152). Traces of tin are also present within the interior of the pommel cap, which could indicate the use of solder by which the rivet was fixed to the internal apex of the cap (Petri 2017: 151–152). On both sides between the tang, the surface of the upper guard is perforated by circular holes that served to accommodate the rivet (Petri 2017: Pl. 5a). The pommel cap is no longer present on the sword today, which is also confirmed by the description in the SMB catalogue. The ornamentation of the hilt components has not survived. However, the presence of faint grooves observed on both the upper guard and the pommel cap, as well as preserved residues of white metal (identified as tin), suggest that the pommel and crossguard may have been covered with a layer of non-ferrous metal in the past (Petri 2017: 152–153).
On one side of the blade, the inscription +VLFBERH has been preserved, along with a faintly visible equilateral cross and a vertical line following the letter H, which the X-ray radiograph reveals to be the letter T (Petri 2017: 152–153, Fig. 10a, 11). The inscription on the blade could thus belong to the first variant of the inscription group according to Anne Stalsberg (+VLFBERH+T), despite the fact that the researcher classified this inscription under Variant 6 (Stalsberg 2008: Cat. No. PL1). The reverse side is covered by fragments of a geometric mark—most likely a lattice pattern accompanied by vertical lines (Petri 2017: 152–153, Fig. 10b, Pl. 6b). Morphologically as well as metrically, this sword may be close to Geibig’s combinational Type 11 (Petri 2017: 155). Ingo Petri comparatively correlates it with the sword from Haddebyer Noor, which is designated by some authors as a Petersen Type U based on the low proportions of its pommel (Petri 2017: 156). One cannot agree with this assertion, as the structural parts of the hilt exhibit decoration typical of Jan Petersen Type V swords featuring a chequered pattern (Geibig 1989: Cat. No. 27, Pl. 6.1). The author further notes that although the sword from Awajki is classified under Petersen Type T2 in archaeological literature, he highlights that the morphological attributes of the pommel cap do not display the characteristics of the type under study (Petri 2017: 156–157). In his view, the sword is therefore difficult to assign to a specific type; he also suggests a possible monochrome decoration as well as manufacture by a local artisan (Petri 2017: 157–158).
Fig. 47. The sword from Awajki (Awecken): a1–2. close-up of the structural components of the hilt and view of the complete sword, photographs from the estate of Rudolf Grenz (after Petri 2017: Pl. 5b);
b. view of the current condition of the sword (photograph adapted from recherche.smb.museum – Staatliche Museen zu Berlin, Museum für Vor- und Frühgeschichte).
Summary
Based on our observations focused on morphological and decorative attributes, out of the total number of 42 swords analysed above, we infer that 27 swords belong to the type under study, while another 11 represent specimens with a possible typological classification under this group. In our opinion, the remaining four swords and their structural components (Gramzdas Upītes, Kokneses kapulauks, Turaidas Pūteļi, Awajki) are difficult to assign to a specific type. Although the crossguard of the first two and the upper guard of the second specimen convey the impression of the type under study, the absence of a pommel cap and the unknown condition of the decoration (if indeed any has survived) prevent us from drawing a satisfactory conclusion regarding their typological assignment. Due to the poor quality of the photograph, the third specimen in this sequence cannot be evaluated more effectively. In our view, based on surviving older photographs, the last sword from Awajki is linked to Type T2 solely by the method of fastening the pommel components; however, this method is identical to that found on Type U or Type V swords, which are also characterised by a three-part division of the pommel. The sword from Awajki may therefore also represent a local product imitating one of the aforementioned types (T2, U, V).
Jan Petersen sought the origin of Type T swords with blades bearing Ulfberht inscriptions within the Frankish Empire, though he considered the origin of the type as a whole to be uncertain (Petersen 1919: 153). The highest frequency of occurrence is recorded in Old Rus’, followed by Scandinavia and the Baltic Sea region, but this is insufficient to place the origin of the type precisely within these areas (Maczek et al. 2024: 294). Several researchers are united in the view that the origin of the type must be sought outside the territory of the Carolingian Empire, although precise localisation is currently difficult to determine (Martens – Astrup 2021: 134; Maczek et al. 2024: 294), a stance with which we fully concur. Even though not a single grave find of Type T2 swords – nor of others such as Types S, T1, or V – is known from the environment of the former Holy Roman Empire, the royal courts and elites stood relatively close to them, as such swords frequently became coronation objects or relics (Schulze-Dörrlamm 2014: 610, 612, 619, Fig. 1.1–2; Viskupič 2025).
In our estimation, the shape of the pommels is related to Type E as well as Types T1, U, and V; however, in contrast to the T1 variant, the structural components of the hilt exhibit neither massiveness nor a prominently shaped central segment. The decoration, nonetheless, links this variant with Type V. Due to the morphological inconsistency of the pommels, some authors view a portion of the analysed swords as transitional or hybrid variants between both types (Maczek et al. 2024: 291–296). Certain specimens, such as the sword from Vakhrushevo with its rounded central segment and less pronounced bridges of the central segments, may indeed represent a transitional type where the pommel is closer to Type V, but the decoration shifts towards Type T2.
The dating also appears complex, given that not all grave finds have a clear archaeological context, and the remaining swords and their structural components constitute stray or river finds. The type itself is represented by only a small number of specimens, which further complicates its more detailed analysis.
Generally, Type T2 is placed within the period from the second half of the 10th century to the beginning, or alternatively the first half, of the 11th century (Petersen 1919: 152, 183; Moilanen 2015: 253). Kirpichnikov attributes the type to the 10th century (Kirpichnikov 1966a: 30, Cat. Nos. 39–42). Fedir Androshchuk inclines towards a dating within the Late Viking Age, specifically between the years 970 and 1000 (Androshchuk 2013: 159), or shifts the boundary to between 975 and 1050, arguing their absence from Birka (Androshchuk 2014: 172). For only a very small number of them can we assume that they certainly did not appear before the year 950, which, however, does not provide a clear picture of the period of use for the Type T2 variant as a whole.
Appendix
The link below to the tabular format of the Petersen Type T2 sword catalogue contains detailed information related to individual sword specimens, such as metric data, inventory numbers, find contexts, types of decoration used, inscriptions and marks on the blades, etc.
Acknowledgements
At the very conclusion of this paper, I would like to express my warmest gratitude to Tomáš Vlasatý for his valuable consultations and for providing the opportunity to publish my work.
Bibliography
Aannestad, H. L. (2018). Charisma, Violence and Weapons. The Broken Swords of the Vikings. In: Vedeler, Marianne et al. (eds). Charismatic Objects: From Roman Times to the Middle Ages, 147–166.
Androschchuk, Fedir (2014). Viking Swords: Swords and Social aspects of Weaponry in Viking Age Societies, Stockholm.
Androshchuk 2001 = Андрощук, Ф. А. (2001). Гнездово, Днепровский путь и финал Бирки. In: Гнездово: 125 лет исследования памятника. Труды Государственного исторического музея, Москва, 126-135.
Androshchuk 2013 = Андрощук, Федор (2013). Мечи викингов, Київ.
Androshchuk – Zotsenko 2012 = Андрощук, Фёдор – Зоценко, Владимир (2012). Скандинавские древности Южной Руси: каталог (Scandinavian Antiquities of Southern Rus´), Paris.
Atgāzis, Māris (2019). Tuvcīņas ieroči senajā Latvijā 10.–. 13. gadsimtā, Rīga.
Baxa, Peter (1974). Nález včasnostredovekého meča na Slovensku. In: Archeologické rozhledy 26, 60–62.
Brandenburg 1895 = Бранденбург, Н. Е. (1895). Курганы Южного Приладожья. In: МАР. № 18. Спб, Санкт-Петербург.
Brunning, Sue (2015). ‘(Swinger of) the Serpent of Wounds’: Swords and Snakes in the Viking Mind. In: Williams, J. T. (ed.). Representing Beasts in Early Medieval England and Scandinavia, Woodbridge, 53-72.
Budvydas, Ugnius (2007). Аbout Some Aspects of Scalvian Armaments, on the Basis of Investigations in Viešvilė. In: Archaeologia Baltica 8, 205-213.
Cleve, Nils (1978). Skelettgravfälten på Kjuloholm i Kjulo; Vikingatid och korstågstid. Gravfältet C. Finska Fornminnesföreningens Tidskrift 44: 2, Helsinki.
Chernyshev 1963 = Чернышёв, Н. А. (1963). О технике и происхожде нии «франкских» мечей, найденных на Днепрострое в 1928 г. In: Скандинавский сборник VI, 211-228.
Duczko, Wladyslaw (2004).Viking Rus – Studies of the Presence of Scandinavians in Eastern Europe, Leiden – Boston.
Ebert, Max (1914). Ein Schwert mit tauschierte Klinge von Lümmada auf Oesel. In: Baltische Studien zur Archäologie und Geschichte, Berlin, 147–158.
Frait, Radovan (2006). Mladohradištní meč z Univerzitní ulice v Olomouci. In: Zprávy Vlastivědného muzea v Olomouci – společenské vědy, čís. 292, Olomouc 2006, 97–102.
Geibig, Alfred (1989). Zur Formenvielfalt der Schwerter und Schwertfragmente von Haithabu. In: Offa 46, 223-267.
Geibig, Alfred (1991). Beiträge zur morphologischen Entwicklung des Schwertes im Mittelalter, Neumünster.
Geibig, Alfred (1999). Die Schwerter aus dem Hafen von Haithabu. In: Berichte über die Ausgrabungen in Haithabu 33, Neumünster, 9-99.
Goßler, Norbert (2013). Die mittelalterlichen Steigbügel aus dem Berliner Bestand der Prussia-Sammlung (ehemals Königsberg / Ostpreußen) – Studien zu Typologie, Chronologie und Kulturgeschichte. In: Acta Praehistorica et Archaeologica 45, 109–215.
Goßler, Norbert – Jahn, Christoph (2019). Wikinger und Balten an der Memel: Die Ausgrabungen des frühgeschichtlichen Gräberfeldes von Linkuhnen in Ostpreußen 1928–1939, Mainz.
Hjardar, Kim – Vike, Vegard (2016). Vikings at War, Oxford – Philadelphia.
Hošek, Jiří – Košta, Jiří – Žákovský, Petr (2021). Ninth to mid-sixteenth century swords from the Czech Republic in their European context. Part II, Praha – Brno.
Jahn, Christoph – Ibsen, Timo (2024). Wiskiauten/Mochovoe: Das wikingerzeitliche Gräberfeld: Funde und Befunde der Ausgrabungen 1865−1939, Baden-Baden.
Jakobsson, Mikael (1992). Krigarideologi och vikingatida svärdstypologi, Stockholm: Stockholms Universitet.
Janionis, Andrius (2025). V–XV a. kalavijų raida Lietuvos archeologinės medžiagos archeometalurginių tyrimų kontekste, Vilnius: Vilniaus universitetas.
Jets, Indrek (2013). Lahingu maod: Skandinaavia 9.–11. sajandi kunstistiilid Eesti arheoloogilistel leidudel, Tallinn.
Kainov, S. Yu. (2012). Swords from Gnёzdovo. In: Acta Militaria Mediaevalia VIII, 7-68.
Kainov 2019a = Каинов, С. Ю. (2019). Сложение комплекса вооружения Древней Руси X – начала XI в. (по материалам Гнёздовского некрополя и поселения). Диссертация на соискание ученой степени кандидата исторических наук Том I, Москва.
Kainov 2019b = Каинов, С. Ю. (2019). «Большой» меч из Чёрной могилы (предварительные итоги нового этапа изучения). In: Земля наша велика и обильна : сборник статей, посвященный 90-летию А. Н. Кирпичникова. Отв. ред. С.В. Белецкий, Санкт-Петербург, 125-139.
Kainov – Novikov 2024 = Каинов, С. Ю. – Новиков, В. В. (2024). Новые находки мечей в Гнёздове (2017-2020 гг.), Москва.
Kainov – Shchedrina = Каинов, С. Ю. – Щедрина, А. Ю. (2025). Меч из коллекции Историко-художественного музея Рославля (Смоленская обл.). In: Археологические вести, 48, 227-242.
Kazakevičius, Vytautas (1996). IX–XIII a. baltų kalavijai, Vilnius.
Kåresen, Beate Krydsby (2024). A Functional Analysis of the Rattles and Hooks of the Viking Age, Oslo : University of Oslo.
Kirpichnikov 1966a = Кирпичников, А. Н. (1966). Древнерусское оружие: Вып. 1. Мечи и сабли IX–XIII вв, Москва – Ленинград.
Kirpichnikov 1966b = Кирпичников, А. Н. (1966). Древнерусское оружие. Вып. 2: Копья, сулицы, боевые топоры, булавы, кистени IX – XIII вв, Москва – Ленинград.
Kirpichnikov 1992 = Кирпичников, А. Н. (1992). Новообнаруженные клейма раннесредневековых мечей. In: Fasciculi Archaeologiae Historicae. Fasc. V, 61–81.
Kirpichnikov, A. N. – Stalsberg, Anne (1998). New investigations of Viking Age swords materials in Norwegian museums. In: Wesse, Anke (ed.). Studien zur Archäeologie des Ostseeraumes. Von der Eisenzeit zum Mittelalter, Neumünster, 507-514.
Kiudsoo, Mauri (2019). Viikingiaja Eesti: ajakirja Imeline Ajalugu eriväljaanne, nr. 1, Tallinn.
Kochkurkina 1989 = Кочкуркина, С. И. (1989). Памятники Юго-Восточного Приладожья и Прионежья X–XIII вв., Петрозаводск.
Kochkurkina – Summanen 2021 = Кочкуркина, С. И. – Сумманен, И. М. (2021). Средневековые древности. Каталог археологических коллекций ИЯЛИ КарНЦ РАН, Петрозаводск.
Komar 2014 = Комар, А. В. (2014). Мечи Днепростроя (к истории находки 1928 г.). In: Русь в IX–XII веках: общество, государство, культура, Москва – Вологда, 47-61.
Košta, Jiří (2021). Early Medieval Swords from the Czech Republic (9th–10th Centuries), Praha.
Kotowicz, P. N. (2018). Early Medieval Axes from Territory of Poland, Kraków.
Kovač, Mario (2003). Ubojite oštrice – Hladno oružje na području Hrvatske od IX. do kraja XVIII. stoljeća, Gornja Stubica.
Kovalenko et al. 2020 = Коваленко, Е. С. et al. (2020). Рентгеновская, синхротронная и нейтронная визуализация металлических артефактов из кургана Черная Могила. In: Российские нанотехнологии, том 15, № 5, 56–68.
Kulakov 1989 = Кулаков, В. Н. (1989). Знамёна дружин Балтийского региона. In: Советская археология 4, 61-70.
Kulakov 2019 = Кулаков, В. И. (2019). Декор рукоятей мечей эпохи викингов из юго-восточтной Балтии. In: Брянского государственного университета, вып. 4, Брянск, 74-83.
Leppäaho, Jorma (1952). Constantinus Rex. In: Corolla Archaeologica in honorem C. A. Nordman, 145–150.
Maczek, Dušan et al. (2024). Early Medieval Sword from Košice-Krásna revisited. In: Zborník Slovenského národného múzea CXVIII, Archeológia 34, 277–301.
Mandel, Mati (1991). Eesti 8.–13. sajandi mõõkade tüpoloogiast ja dateerimisest. In: Muinasaja teadus 1, 101–133.
Martens, Irmelin (2009). Vågå og Tinn – to bygder med mange praktsverd fra vikingtid. In: Viking 72, 183-196, Oslo.
Martens, Irmelin – Astrup, Eva E. (2021). Viking Age Swords from Telemark, Norway. An Integrated Technical and Archaeological Investigation, Oslo.
Mägi, Marika (2017). Viikingiaegne Eesti, Tallinn.
Moilanen, Mikko (2015). Marks of fire, value and faith : swords with ferrous inlays in Finland during the late Iron Age (ca. 700-1200 AD), Turku.
Moilanen, Mikko (2018). Viikinkimiekat Suomessa, Helsinki.
Môc, Róbert (2018). Príspevok k technológii výroby meča z Krásnej nad Hornádom. In: Zborník Slovenského Národného Múzea 112, 221-233.
Mühlen, Bernt von Zur (1975). Die Kultur Der Wikinger in Ostpreussen, Bonn.
Müller-Wille, Michael (1970). Ein neues Ulfberht-Schwert aus Hamburg. Verbreitung, Formekunde und Herkunft. In: Offa 27, 65-91.
Murasheva, V. V. et al. (2021). “Barbarian Scepters” of the Viking Age from Chernaya Mogila burial mound at Chernigov (present-day Ukraine). In: Journal of Archaeological Science: Reports 37, 1-11.
Paulsen, Peter (1953). Schwertortbänder der Wikingerzeit: ein Beitrag zur Frühgeschichte Osteuropas, Stuttgart.
Pedersen, Anne (2014). Dead Warriors in Living Memory. A study of weapon and equestrian burialsin Viking-age Denmark, AD 800-1000.Publications from the National Museum. Studies in Archaeology & History Vol. 20:1 1. (Text), Copenhagen.
Petersen, Jan (1919). De Norske Vikingesverd: En Typologisk-Kronologisk Studie Over Vikingetidens Vaaben, Kristiania.
Petersen, Jan (1951). Vikingetidens redskaper, Oslo.
Petri, Ingo (2017). Vier VLFBERHT-Schwerter aus der Sammlung des Museums für Vor- und Frühgeschichte, Staatliche Museen zu Berlin. In: Acta Praehistorica et Archaeologica 49, 137–187.
Pienimäki, Aleksi (2016). Desalination of iron. A comparison of fresh, wet finds and dry, storage ones, Helsinki : Helsinki Metropolia University of Applied Sciences.
Ravdonikas 1933 = Равдоникас, В. И., (1933). Надписи и знаки на мечах из Днепростроя. In: Известия Государственной академии истории материальной культуры. № 100, 598–616.
Ruttkay, Alexander (1975). Waffen und Reiterausrüstung des 9. bis zur ersten Hälfte des 14. Jahrhunderts in der Slowakei I. In: Slovenská archeológia 24/1, 119–216.
Ruttkay, Alexander (1976). Waffen und Reiterausrüstung des 9. bis zur ersten Hälfte des 14. Jahrhunderts in der Slowakei II. In: Slovenská Archeológia 24/2, 1976, 245-395.
Rybakov 1949 = Рыбаков, Б. А. (1949). Древности Чернигова. In: МИА. № 11, Ленинград.
Rygh, Oluf (1885). Norske oldsager, ordnede og forklarede, Christiania.
Samokvasov 1916 = Самоквасов, Д.Я. (1916). Могильные древности северянской Черниговщины. Москва.
Schulze-Dörrlamm, Mechthild (2014). Schwerter des 10. Jahrhunderts als Herrschaftszeichen der Ottonen. Zu den Vorläufern des Reichsschwerts und zu dessen Imitationsformen. In: Jahrbuch des Römisch-Germanischen Zentralmuseums Mainz, Bd. 59 Nr. 2, 609-651.
Sénby Posse, Lovisa (2021). En rumslig analys av båtgravskicket. Tvärvetenskapliga metoder för att tolka Valsgärdes tvillinggravar 12 och 15, Uppsala: Uppsala Universitet.
Sizov 1902 = Сизов, В. И. (1902). Курганы Смоленской губернии. Гнёздовский могильник близ Смоленска. In: Материалы по археологии России 28, Санкт-Петербург.
Solberg, Bergljot (1984). Norwegian Spear-Heads from the Merovingian and Viking Periods, Bergen : Universitetet i Bergen.
Stalsberg, Anne (2008). Herstellung und Verbreitung der VLFBERHT-Schwertklingen. Eine Neubewertung. In: Zeitschrift für Archäologie des Mittelalters 36, 89-118.
Shchedrina 2018 = Щедрина, А. Ю. (2018). Новый взгляд на де таль меча из коллекции Харьковского историче ского музея. In: Харьковский историко-археологический сборник 23, 18-32.
Shchedrina 2021 = Щедрина, А. Ю. (2021). Два средневековых меча из собрания НМРК. In: Вестник Национального музея Республики Карелия 8, 270-277.
Shchedrina 2023 = Щедрина, А. Ю. (2023). Оружие Приладожских курганов из раскопок И.П. Крупейченко (коллекция Тихвинского музея). In: Воинские традиции в археологическом контексте: от позднего Латена до позднего Средневековья, Тула, 92-113.
Takhtay 1935 = Тахмай, A. (1935). Франкский меч Херсонского музея. In: Проблемы истории докапиталистических обществ, № 5-6, 130-133.
Thunmark-Nylén, Lena (1998). Die Wikingerzeit Gotlands II: Typentafeln, Stockholm.
Thunmark-Nylén, Lena (2006). Die Wikingerzeit Gotlands III: 1–2 : Text, Stockholm.
Tomanterä, Leena (1972). Kolme Köyliön Vanhankartanon miekkaa. In: Suomen Museo 79, 19-28.
Tomsons, Artūrs (2018). Zobeni Latvijas teritorijā no 7. līdz 16. gadsimtam, Rīga.
Viskupič, Michal (2025). Petersen Type V Sword. In: Project Forlǫg – Reenactment and science [online]. [2026-06-13]. Available from: https://sagy.vikingove.cz/en/petersen-type-v-sword/
Viskupič, Michal (2026a). Petersen Type T Swords: Part I – Type T1. In: Project Forlǫg – Reenactment and science [online]. [2026-06-13]. Available here: https://sagy.vikingove.cz/en/petersen-type-t-swords-part-i-type-t1/
Viskupič, Michal (2026b). Early Medieval Double-Edged Swords from the Territory of Serbia (9th–11th Century). In: Project Forlǫg – Reenactment and science [online]. [2026-06-13]. Available here: https://sagy.vikingove.cz/en/early-medieval-double-edged-swords-from-the-territory-of-serbia-9th-11th-century/
Vlasatý, Tomáš (2016A). The helmet from Lokrume, Gotland. In: Project Forlǫg – Reenactment and science [online]. [2026-06-13]. Available here: https://sagy.vikingove.cz/en/the-helmet-from-lokrume-gotland/
Vlasatý, Tomáš (2016B). Sekery z Birky. In: Project Forlǫg – Reenactment and science [online]. [2026-06-13]. Available here: https://sagy.vikingove.cz/sekery-birka/
Vlasatý, Tomáš (2020a). The Biography of a Sword. In: Project Forlǫg – Reenactment and science [online]. [2026-06-13]. Available here: https://sagy.vikingove.cz/en/the-biography-of-a-sword/
Vlasatý, Tomáš (2020b). Wire wrapped sword grips of 9th-11th century. In: Project Forlǫg – Reenactment and science [online]. [2026-06-13]. Available here: https://sagy.vikingove.cz/en/wire-wrapped-sword-grips-of-9th-11th-century/
Vlasatý, Tomáš (2021). Long knife from the grave Valsgärde 12. In: Project Forlǫg – Reenactment and science [online]. [2022-12-08]. Available here: https://sagy.vikingove.cz/en/long-knife-from-the-grave-valsgarde-12/
Vlasatý, Tomáš (2022a). Latvian metal wrapped axe shafts. In: Project Forlǫg – Reenactment and science [online]. [2026-06-13]. Available here: https://sagy.vikingove.cz/en/latvian-metal-wrapped-axe-shafts/
Vlasatý, Tomáš (2022b). Supplement to Androshchuk’s sword catalog of Viking Age Sweden. In: Project Forlǫg – Reenactment and science [online]. [2022-04-19]. Available here: https://sagy.vikingove.cz/en/supplement-to-androshchuks-sword-catalog-of-viking-age-sweden/
Vlasatý, Tomáš (2026). Swords with Double Fullers: 8th–12th Centuries. In: Project Forlǫg – Reenactment and science [online]. [2026-06-13]. Available here: https://sagy.vikingove.cz/en/swords-with-double-fullers-8th-12th-centuries/
Volkaitė-Kulikauskienė, Regina (1964). IX-XII amžių kalavijai Lietuvoje. In: Iš lietuvių kultūros istorijos IV, Vilnius, 197–226.
Wikell, Roger (2019). Vikingar i norra Haninge. In: Glimtar från Haningebygden 38, 2-5.
Williams, Alan (2012). The Sword and the Crucible, Leiden – Boston.
Zariņa, Anna (2006). Salaspils Laukskolas kapulauks. 10.-13. gadsimts, Rīga.