Research Paper

Through the Portal: Viking Motifs Incorporated in the Romanesque Style in Telemark, Norway

Author: Kristine Ødeby (University of Oslo)

  • Through the Portal: Viking Motifs Incorporated in the Romanesque Style in Telemark, Norway

    Research Paper

    Through the Portal: Viking Motifs Incorporated in the Romanesque Style in Telemark, Norway



This paper presents the results of an analysis of motifs identified on six carved wooden Romanesque portal panels from the Norwegian county of Telemark. The findings suggest that animal motifs in the Late Viking style survived long into the Late Medieval period and were reused on these medieval portals. Stylistically, late expressions of Viking animal art do not differ a great deal from those of the subsequent Romanesque style. However, their symbolical differences are considered to be significant. The motifs themselves, and the issue of whether the Romanesque style adopted motifs from pre-Christian art, have attracted less attention. The motif portraying Sigurd slaying the dragon is considered in depth. It will be suggested that Sigurd, serving as a mediator between the old and the new beliefs when he appeared in late Viking contexts, was given a new role when portrayed in Christian art. Metaphor and liminality are a central part of this paper, and the theories of Alfred Gell and Margrete Andås suggest that the portal itself affects those who pass through it, and that the iconography is meaningful from a liminal perspective.

Keywords: Liminality, Symbol, Portal, Romanesque, Motif

How to Cite:

Ødeby, K., (2013) “Through the Portal: Viking Motifs Incorporated in the Romanesque Style in Telemark, Norway”, Papers from the Institute of Archaeology 1(1), Art. 15. doi:



Published on
24 Sep 2013
Peer Reviewed


The Romanesque style was carved on stave church portals and other Christian art objects in Norway from the 12th century onwards. It was a composite of late Viking art and European impulses (Anker 1998: 135–136; Fuglesang 1981; Hohler 1999a). The ornamentation on stave church portals parallels that on the portals of secular buildings from the 13th and 14th centuries.

The coming of Christianity brought a wave of new styles and motifs to Norway which was adapted by woodcarvers, and which can be seen in the artistic carvings in stave churches and on medieval furniture (Anker 1997: 222). In Norway’s deepest valleys, tradition outweighed renovation. Therefore it is interesting to take a closer look at the use of motifs in this new church style - the Romanesque style. It is relevant to ask whether the Viking motifs were deeply ingrained in the woodcarver’s repertoire.

The overarching question of this research concerns reuse of motifs from Norse animal art in the Romanesque style as seen on wooden portals. By analysing the use of motifs on six wooden portals from the Norwegian county of Telemark, I will attempt to answer the following specific questions: Which motifs from late Norse animal art are reused in the Romanesque style, and how can they be interpreted? Which motifs are used on the stave church portals compared to the profane portals? To what degree can the motifs be interpreted as symbols? And lastly, in the medieval period, was the portal considered not only as a metaphor, but also in relation to the bodies that passed through it? This last question will be addressed first as I explore the portals’ metaphorical function by applying theories of liminality and ‘the enchantment of technology’. Next, results from the analysis of the portals’ individual motifs will be presented followed by a discussion of their possible symbolical and metaphorical interpretation. Lastly, a closer look will be taken at why different motifs appear in different contexts.

Medieval portal ornaments have been researched previously by art historians, architects and building historians, but have traditionally not been a field of study for archaeologists. Other fields of study have also considered the morphological and iconological aspects of medieval art (Hansen 1995: 12). However, I would argue that this project is an archaeological project. First, the Norse animal art of the late Iron Age has been studied mainly by archaeologists (see Domeij 2004; Hedeager 1999b; Hedeager 2004; Klæsøe 2002; Kristoffersen 2000, 2010; Salin 1935). The reason for this is that the Norwegian Iron Age is classified as a pre-historic period. The comparison of material from different periods is not unusual in archaeology and is a key aspect of this paper. Second, the use of metaphorical thinking, such as the phenomenological aspects surrounding the experience of entering through a portal, and the ethnographic and anthropological aspects of liminality are widely used in archaeology. Thus, although it rarely is, studying portal ornaments should be interdisciplinary.

Previous Research

Stave churches and objects connected to them have received special attention due to their unique place in European architectural history, and their symbolic place in Norway’s nationalistic development during the Romantic period. Unfortunately, most of the churches have been demolished. Aside from the portals themselves, sources for the portals stretch back to the mid-19th century. Before AD 1860 churches and portals, which have since been demolished or badly weathered, were sketched by architects (Hohler 1999a: 9). Art historians later studied these sources and published the sketches, among them Dietrichson (1892, 1902), Fett (1909) and the archaeologist Nicolaysen (1853–55, 1861, 1903). A number of recent works also depict stave churches and their ornaments. Work on stave churches by Hauglid (1969, 1973, 1976), Fuglesang (1980, 1981, 1982) and Anker (1997) connect the woodcarving to Viking ornaments. Romanesque expression in the carvings was given attention by Blindheim (1965) and Hohler (1999a). Hohler documented the entire collection of stave church sculpture, including all remaining portals and fragments of portals.

Secular portals have been similarly treated. Berg (1991, 1993) wrote an overview of all medieval secular buildings in Norway, and Gjærder (1952) was the first to document all secular portals. This paper builds on previous research and documentation, but takes it further by schematically identifying portal motifs and comparing them to the motifs used in the late Viking style.

The Portal as Part of a Liminal Zone

The key to historic doors - the key to portals in this context - can be obtained from metaphors. The object itself must be taken into account in the study of art. As Gell’s theory of ‘the enchantment of technology’ states: the power technical processes can fascinate the viewer with allows the viewer to take an enchanted look at the real world (1999: 163). The technology of enchantment, on the other hand, demonstrates a certain level of technical excellence that only a work of art that is made beautiful can achieve. Art can thus be understood as a component of technology.

Andås (2007) explores the relationship between space and ritual. She chooses to describe the physical space around the church as a ‘liminal zone’; an area between the consecrated and non-consecrated ground. She describes how the door played an important role in the ritual of public penance in the early 13th century (Andås 2007: 47). Rituals, physical location and movements were clearly related to its meaning. The door constituted a space ‘in-between’, neither inside nor out, and the rituals that took place at the portal had a between-ness to them. The rituals themselves were focused on the input or transformation by recalling a biblical transformation or highlighting a significant moment in an individual’s life.

As a final comment, Andås (2007: 123) emphasises that the choice of iconography in the middle of the 13th century tells us something essential about the cathedral with regard to:

… what was considered important at the time, about their religious and political ambitions, about how they perceived their political reality and about their strategies. As the religious centre of a large province, the cathedral was a communicator of ideas and political attitude. The Cathedral was the architectural key monument of its province. The construction of something as prominent as the south chancel porch, would not have gone unnoticed. The message communicated would undoubtedly soon have been observed by other political agents of the Norwegian Medieval state. It is in this light that the iconography should be understood.

Iconography should therefore be understood in the light of the building’s function in society. With regard to Nidaros cathedral, it was a key architectural monument and religious centre for its district, and thus an important communicator of ideas and political attitudes. The construction of the choir’s south entrance would have been noticed, and the iconographical message would have been noted by other political agents. The quote above proposes two thoughts relating to this project’s analysis. First, it proposes that the stave church portals can be interpreted in the same light and second, the iconography and motifs could have a far greater political role than previously thought. If this is the case, the iconography of the liminal zone had a direct influence on the rituals outside the church, and possibly actions which took place in the courtyard outside a secular storehouse.

Andås’ theories can be adapted to stave church portals. The church was built on consecrated ground. Whether it was a cathedral or a parish church, or if it was built in wood or stone, probably had little impact on the portal’s status or metaphorical meaning. The portal would have led people into God’s house. According to Andås (2007: 84) the doorway represents a threshold of the eternal law and the manifestation of the sacred law as a guide to human life. The church cemetery hosts both sacred and secular jurisdiction, as seen in the local courts held in the cemetery.

Whether portals in storehouses can be assigned to the same trait must also be discussed. The main physical difference between the church portals and profane portals is that profane portals were much lower and wider. As a result, a person would have to bow his/her head and walk sideways through the portal to enter. This indicates that entering through a profane portal was a completely different experience compared to entering through a church portal. Through the portal went a connection between home and the outside world. A closed door was a symbol of private law (Gjærder 1952: 11). Also, the medieval profane doors were heavy, so it was a process in itself just to open them (Gjærder 1952: 226). The storehouse, a stabbur (Figure 1), had a specific function as a storage compartment. The storage of food had to be indoors, where animals and uninvited guests had no access. The door was locked and only the owner had the key. Medieval stabbur were independent treasuries with one or two floors, to store grain, flour, bread, fish and meat products. On the upper floor was a grand bedroom for guests and storage for clothes, it was called a loft (Berg 1989: 174). Christensen (1998: 265) writes that the living conditions in the medieval period were largely controlled by magical performances; at the entrance to the barn or storehouse, tar crosses and six leaf roses were painted and there could be inscriptions that asked for blessings or wood carved masks on the doors. Gjærder (1952: 31) states that the doors of the medieval log houses were largely without the complicated ornamentation seen on stave churches, but showed far more types and varieties, and a larger range of decorative motifs.

Figure 1
Figure 1
Figure 1

The storage house from Tveito, Telemark, in its new environment at the Norwegian Folk Museum in Oslo. The ground floor portal and its three-dimensional carved capital lions can be glimpsed from a distance. Photo: author.

The 12th century manuscript Gamal Norsk Homiliebok, contains spiritual statements relating to different parts of a church building. One part of the manuscript, the Stave Church Sermon, could have been used once a year as a reminder of the church’s consecration (Hjelde 1990: 290). In the Stave Church Sermon the portal is described as a metaphor. For the people who used the portal in everyday life, it may also have been seen as a metaphor. The portals practical feature was as an entry and exit to the building. It also marked a liminal zone and physical and bodily experiences were important aspects of its use. The body is the most important element involved in actions. The portal has no meaning without people to see it, touch it and go through it. Anyone who enters the building must pass through it and finds themselves surrounded by the portal: welcoming in front when outside, towering over when standing on the threshold and protecting from behind when inside. When the portal has such a significant task as the building’s entrance, and is significant to the consciousness of the person who walks through it, the portal has a physical aspect. The world is experienced with the body and human practice makes the world available to us through a particular perception of it. It directs us as spectators by constructing a relationship between us and the things around us (Fredriksen 2005: 210).

The impact of the motifs can start long before people go through the portal. The major motifs, especially the capital animals were visible from a distance. Among the portals surveyed in this analysis the Nissedal portal is the only one currently in its original location (see Table 1). The motifs on the Nissedal portal were likely to have been highly visible when the portal was new and visible from some distance. Today, the wood is decomposing and it is difficult to distinguish the different motifs, especially from afar. The portal of Southern Tveito is in the Tveito loft at the Norwegian Folk Museum (Figure 1), and is better preserved. It has capital animals and a three-dimensional carved mask. The capital animals can be seen at a distance. From 10 to 20 m away one can make out the motifs and recognise the animal motifs. At a distance of 40 to 60 m one can see that the capitals and side pieces are ornamented, but not what they represent.

Table 1

The analysed portals

Portal Original location Dating
The Sauland portal Sauland stave church in Hjartdal, Telemark Stylistically dated to the late 12th or early 13th century.
The Nesland portal Unknown. Reused in Nesland stave church in Nystu søndre, Vinje, Telemark Stylistically dated to ca. AD 1200. According to a runic inscription the church was consecrated in AD 1242.
The Lisleherad portal Lisleherad stave church in Landsverk nordre, Heddal, Telemark Stylistically dated to the 13th or early 14th century.
The Tveito portal The farm Søre Tveito in Hovin, Tinn, Telemark Dated to ca. AD 1300 based on style, a runic inscription and the log construction.
The Lundevall portal The farm Lundevall in Kviteseid, Telemark Stylistically dated to AD 1250–1350.
The Nissedal portal The rectory in Nissedal, Telemark Prior to AD 1350 based on the construction. Previously the portal has stylistically been dated to the early
13th century.

To distinguish between the different motifs, the carving may have been painted. The Nesland portal has remnants of blue and white paint on the columns (Figure 5) but it is unlikely that this is original. The west portal from the nave of Flesberg stave church, Buskerud is painted with blue, yellow and red but the paint is newer than the portal. Meanwhile, the west portal from Nore stave church, Numedal is painted with green, yellow and red. Gjærder (1952: 30) states that the medieval profane doors were also coloured. On the Tveito portal there remains original ochre and rust red pigment. Today the colours are heavily worn and eroded in some places, but enough is preserved to establish that polychrome played a role in portal art. Polychrome was used as an accompaniment to the carvings, and seemed to separate motifs from each other (Gjærder 1952: 30). Thus the motifs on both ecclesiastical and profane portals could be seen from a distance, and were not necessarily just for those who went through the portal.

Figure 2
Figure 2
Figure 2

The Sauland portal, depicting the classic Sogn-Valdres design. Photo: John Erling Blad.

Figure 3
Figure 3
Figure 3

The Lundevall portal portraying Sigurd slaying the dragon. Drawing: P. Gjærder (1952: 63).

Figure 4
Figure 4
Figure 4

Inhabited vine on the Sauland portal. Photo: author.

Figure 5
Figure 5
Figure 5

The topmost motif on the Nesland portal: The Creation, depicting a standing halo-wearing God creating Adam and Eve. Photo: author.

In over half of the extant stave churches, the carved portals are not visible from the outside of the church. These portals are usually set into a supporting wall inside the built galleries. The portals in these cases had a different effect since they could not be seen from afar and one could prepare to meet them. An example can be seen in Heddal stave church. Here the four portals are unlit in dark galleries. They are very tall and almost reach the roof. The viewer thus meets the portals suddenly and is overwhelmed. The portals must have been perceived as large and powerful and also magnificent (the technology of enchantment) and magical (the enchantment of technology). Where the portal is experienced in its magnificence, it is stunning because it is a product of technology and ‘made’ beautiful, and has thus achieves the technical level of excellence. Where the portal is perceived as a magical object, medieval people saw the portal in an enchanted light.

Norse Animal Art, the Portals and their Ornaments

In this research, wood was chosen as the research material for the Romanesque style since it could be studied in both ecclesiastical and secular contexts. In Norway the stone portals from secular houses are not preserved. Wood was the main material for building houses in the medieval period since it was cheap and easy to transport. It was also the material used in traditional Viking woodcarving. With the exception of a few halls and fortresses, only churches were built in stone (Ekroll 2000: 13). It is in Norwegian woodcarving that the old ornaments are best preserved, made in a milieu where traditions were kept and impulses held at a distance (Hohler 1984: 145). Within these traditional ornamental styles, it is the stave church carvings that dominate, especially the beautifully composed Sogn-Valdres design (Hohler 1984: 145–146; 1999b; see Figure 2). The woodcarver’s choice of ornaments must, however, have been determined by a limited set of compositions (Hohler 1984: 146).

This analysis includes six portals from the Norwegian county of Telemark, three stave church portals and three portals from secular buildings (Table 1).

Using Panofsky’s iconological method to analyse the portals’ carvings, a number of motifs were identified. Panofsky’s iconological method consists of three levels: pre-iconographical description, iconographical analysis and iconological interpretation. The aim of pre-iconographical description is to find the image’s primary topics by identifying pure forms. The iconographical analysis puts the image in the context of literary sources, and the iconological interpretation tries to understand the artist’s choice and presentation of designs, based on contemporary national principles (Panofsky 1982: 30). Panofsky’s method was significant in identifying the portals’ motifs and their meaning. Table 2 presents the identified motifs from the six analysed portals.

Table 2

Identified motifs from the six analysed portals. The question marks indicate an uncertain interpretation based on literary sources

Portal motifs Specifications and interpretation
Snake Head seen from front
Ribbon-shaped animal Head in profile, 2–4 legs
Dragon Head in profile, legs and wings
Four-legged animal -
Mask Man’s face
Lion’s head -
Bird -
Plant Vine
Horse With a rider, Sigurd’s horse Grani
Battle between dragons and snakes Battle-motif
Battle between four-legged animals Battle-motif
Battle between men and animals Battle motif, Sigurd and Fafnir
Two men and a woman The Creation
Eight people in a boat Noah’s Ark
Three men by the sea The crossing of the Red Sea
Four people Jesus raises a dead girl (?)
Five people Three men in a furnace (?)
Two men David and Goliath

By applying the iconological method the portal motifs were individually identified. Although interpreting motifs in art is not fully objective, this method provides the required level of objectivity. For example, using Panofsky’s three levels, in level one the battle-scene between a large snake and a man is described as the interaction between three figures (Figure 3): a man, a long snake and a horse. The man, wearing a large ring on his back, is running his sword through the body of the snake, which in turn bites the horse’s tail. In level two this scene is cross-referenced with literary sources and analysed to be a monster-slaying hero - possibly St. George, the Archangel Michael, Didrik of Bern or Sigurd the Dragon Slayer. Considering the presence of the ring, the composition most likely depicts Sigurd, whose story revolves around the finger-ring Andvarenaut. In level three, the interpretation reveals that the Sigurd motif is already an old motif when it appears on the Lundevall portal. Stories from the Volsung Legend, where Sigurd originates, were widely used as motifs in Viking styles and the Romanesque style, especially in Norway, Sweden and the Isle of Man. The motif is often found in combination with runic inscriptions, and is often interpreted as an allegory of Christianity’s victory over pagan beliefs. For an elaboration of the interpretation of motifs presented in the tables see Ødeby (2012).

The identification of motifs from the three latest Viking styles are based on other researchers’ results (see Fuglesang 1980, 1981, 1982; Hedeager 1999, 2011; Klæsøe 2002; Kristoffersen 2000; Nielsen and Kristoffersen 2002; Shetelig 1950; Wilson 1980). Only the youngest Viking styles from around AD 950–1100 (Mammen-, Ringerike- and Urnesstyles) were used in the comparison because in those styles Christian motifs can be identified in the ornamentation and they are considered Christian animal styles.

Table 3 presents three categories: motifs exclusively from the Viking styles, motifs exclusively from the Romanesque style and motifs from both. Generally, animal motifs appear in both contexts but there is a noticeable trait relating to the biblical motifs; Old Testament stories are introduced in the Romanesque style and New Testament stories disappear. Aside from this it seems that motifs in the Romanesque style also exist in the Viking styles. The answers to why this is the case can be obtained from discussing the motifs’ symbolic value.

Table 3

Motifs from the late Viking period compared with motifs on the medieval portals

Motifs from the three latest Viking Age styles: Mammen, Ringerike and Urnesstyle Portal motifs
Dragon Dragon
Snake Snake
Ribbon-shaped animal Ribbon-shaped animal
Four-legged animal Four-legged animal
Horse Horse
Bird Bird
Plant Plant
Mask Mask
Sigurd the Dragon Slayer Sigurd the Dragon Slayer
Battle between ribbon-shaped animal and snake
Battle between lion and snake
The Adoration of the Magi
Birth of Christ
The Bethlehem stable
The Creation
Noah’s Ark
The crossing of the Red Sea
Jesus raises a dead girl
Three men in a furnace
David and Goliath
Lion’s head
Battle between two four-legged animals
Battle between dragons and snakes

Symbols in the Norse Animal Art

The symbolic content of the Norse animal art has been studied by a number of scholars (see Domeij 2004; Hedeager 1999; Hedeager 2004; Hedeager 2011; Kristoffersen 2000, 2010). Hedeager (2011: 66) emphasises that the animal’s role should be understood as part of the pre-Christian mindset. The iconographical elements are thought to have been carefully chosen. Art was used in the establishment and legitimisation of power, and thus control and maintenance of style is considered to be an elite strategy (Hedeager 2011: 51).

The complex representations of animals reveal that the style does not reflect the animals themselves, but rather the mentality Iron Age people attributed to them (Kristoffersen 2010: 262). The style expresses a reality more revealing than a naturalistic rendering could have done. Species are mixed together, such as a snake with a bird’s head (Hedeager 2011: 68). In several places complete, or part, human bodies are incorporated. Ambiguousness is an important structuring element, but the semantic potential of animal art is so complex and multidimensional that we cannot possibly grasp its full meaning today. We can, however, draw some conclusions (Hedeager 2011: 73–74). These include that the portrayals cross the boundaries between humans and animals, the depicted animals are wild beasts (not domesticated animals) and that species are portrayed as a reconciliation of animal and human body parts.

The question then arises as to whether medieval motifs can be interpreted as symbols in the same way. Researchers are split into two groups: those who try to find symbolic meaning in the Romanesque style (Anker 1997; Dietrichson 1892; Hohler 1995; Staecker 2006), and those who interpret the design as purely ornamental and without meaning (Anker 1997; Hohler 1981). This question will be discussed in the next section.

Symbols in the Romanesque Style

Arguments for the motifs to incorporate a symbolic content are that both pagan and Christian symbols and symbols of power occur. Fuglesang (1980: 92) identifies the four-legged animals in the Mammen, Ringerike and Urnes styles as lions, and interprets them as royal symbols of power. Half of this project’s material is church portals, and their motifs may have had a symbolic meaning as a portrayal of the Christian church and victory over evil forces (Anker 1997: 267).

An argument against the motifs having a symbolic value is that they may have belonged to a secular portal tradition and that the symbol had lost any meaning. This theory, however, lacks any evidence to suggest that stave church portals had a secular origin and was built on a fully imported compositional scheme (Anker 1997: 267). There are no preserved secular doors in Norway older than ca. AD 1100, but several of these doors show a well-developed style and technical skill in the performance of the craft (Gjærder 1952: 14). This may indicate that the impulses went the other way, that the secular portals drew inspiration from the church portals, when it came to the decor and use of design.

The following is an elaboration of the possible symbolic contents of the Sigurd-motif, the inhabited vine and the general animal motifs.


Sigurd the Dragon Slayer has been interpreted in different ways. In Vǫlsungasaga Odin enters the story, and there are clear references to Norse mythology. Nordanskog (2006: 217) asks whether the motifs were ordered by defiant pagan peasants. However, the portals analysed are from the 12th and 13th centuries when Norway had been Christian for a long time and the story of Sigurd was known and accepted (Nordanskog 2006: 222). According to Blindheim (1972: 11), Sigurd was the ideal of masculinity for the Viking period and medieval men; a hero favoured by gods and loved by women. Sigurd could have symbolised a pride in homely traditions. He was the biggest hero in the Norse sagas, so to use him as a motif was no accident.

Sigurd clearly cannot be connected to Christian beliefs and interpreting secular motifs on church portals is difficult. There would be little point in interpreting them as pagan thus Sigurd cannot be interpreted as either a religious or non-religious motif (Hohler 1995: 171). The Lundevall portal dates to the 13th or 14th century, during which period pagan myths were distant stories. According to Staecker (2006: 366–367) the Sigurd myth played an important part in the transition between pagan and Christian times, acting as a mediator between the old and new beliefs. The question remains why Sigurd is portrayed in ornaments in the late medieval period. There was no further need for a mediator, and his function must therefore have been different.

Dietrichson (1892) suggests that Sigurd represented the ideal Christian knight, and Staecker (2006: 366) views Sigurd as a praefiguratio Christi, or a symbolisation of the dragon slayer St. Michael. In this sense the slaying of the dragon equates to victory over the Devil. Blindheim points out that there is a pseudo-Christian light over the other two main characters in Vǫlsungasaga - Gunnar and Hogne. The knightly ideals may be sensed in the martyrdom of Gunnar and Hogne, and the other protagonist’s revenge. Carved on the choir portal in Lunde Church in Telemark, Sigurd with the dragon and Samson with the lion are portrayed on opposite capitals. When these two myths are paralleled on either side of the church portal, it can explain how the story of Sigurd got a foothold in the church (Blindheim 1972: 20; Nygaard 1996: 131).

The common denominator in the interpretation of Sigurd’s appearance in medieval art is that it represents a subconscious idea of the battle against evil, which offers an explanation of how the design could pass in to an ecclesiastical environment. On the Lundevall portal, one can argue that Sigurd is depicted as a hero. Where he sticks his sword in the snake, the dragon Fafnir, the snake symbolises the Devil. This is underlined by the way poetry was used in church decoration throughout Europe in the medieval period. The story also had a certain historical authenticity. In one version of Ragnar Lodbrok’s saga, it is said that Sigurd’s daughter Aslaug married Ragnar Lodbrok and their son Worm’s-eye, begat Ragnhild - Harald Fairhair’s mother. In this way, Sigurd is an historical person, and genealogically linked to the Norwegian royal family (Hohler 1995: 173; Nordanskog 2006: 221).

Nordanskog (2006: 222) also attempts to explain why Sigurd was carved on churches. The pictures and the story of Sigurd could obviously be placed somewhere it was no problem to use them. It was probably on the stave church portals and the secular portals that our image of the Viking era and the pre-Christian Scandinavian religion was formed. Out of the 40 medieval depictions of Sigurd in Norway, 33 are connected to churches and cemeteries (Blindheim 1972: 20; Liepe 1989: 11). However, ecclesiastical objects are better preserved than secular objects (Blindheim 1972: 20).

Inhabited Vines

Animals and vines are the motifs which art historians have had the greatest difficulty in interpreting. They are not placed into figure or pictorial scenes, but are entangled in each other in larger compositions. This applies to both the Urnes portal, the oldest preserved stave church portal with its famous Urnes style design, and portals with the Sogn-Valdres composition, of which the portal from Sauland is part. Among the most common Romanesque motifs are those which Hohler (1995: 177) calls ‘inhabited vines’, where the animals sit in the vines (Figure 4). The expression itself is thus a part of the Christian Romanesque tradition. However, the combination of three dragons on the lintel over the door opening that is present on the portal from Sauland (see Figure 2), is distinctive for stave church portals and not found on secular portals. Johan Christian Dahl believes that the motifs had a background in Norse mythology, and according to Nicolaysen it may have been a development from the Irish animal ornamentation (Hohler 1995: 177). Dietrichson and Bugge interpreted the vines as the world-tree Yggdrasil, while the struggle between animals symbolised Ragnarok (Hohler 1995: 177). That Ragnarok is the end of the pagan world is underlined by the fact that the motif is placed on the portal to the house of God, and illustrates the final battle between good and evil. This interpretation is now out of date, as is the interpretation of Sigurd’s history as a pagan story, because it is contradictory to interpret pre-Christian symbolism on portals that were made long after the arrival of Christianity.

Animal Motifs

In Christian medieval art individual elements in paintings had their own symbolic meaning or function. Ferguson examined this in Signs and Symbols (1989) in which he established the symbolic meanings of individual motifs including the dragon which symbolises the Devil who is portrayed in Revelations 12:7–9 as God’s enemy. The serpent, often in the same costume as the dragon, is depicted as the one who tempts Adam and Eve in Genesis 3:13. Therefore the serpent represents the general temptation to deceive people and lead them into sin (Ferguson 1989: 17). In the Renaissance, the horse is often depicted as a symbol of desire (Ferguson 1989: 20), which is inspired by Jeremiah 5:8. Birds were used in the early Christian period as symbols of a ‘winged soul’. General, not species-specific, birds were used by artists to represent the spiritual, as opposed to the material (Ferguson 1989: 12). Plants are often more difficult to interpret, but in Renaissance art, the various flowers, trees and plants had different symbolic meanings. Vine elements could be interpreted as grapevines. Vines are one of the most vivid symbols of the Bible, and represent the relationship between God and his people (Ferguson 1989: 39).

Ferguson’s examples from Renaissance paintings illustrate how the symbols were used in Christian art. All of the portals in my analysis are made in, and for, a Christian community. From this it would not be unreasonable to suggest that the symbolic interpretations presented above are real. Previous research, however, rejects this interpretation of portal motifs when the motifs are depicted alongside vines and animals on wooden doors in the Norwegian mountain regions (see Anker 1997; Hohler 1995). Such a fixed symbolic interpretation underestimates, according to my assessment, the complexity of the early Christian community in Norway. Renaissance paintings from the High and Late Medieval period had a different function from portals, and their symbolism cannot be automatically transferred to the designs in such a different context. The portal is more than just art and has a very specific function. Anyone who enters the building must pass through the portal and finds themselves surrounded by it. I suggest that the motifs were selected for the portal to complement this function.

The portals in this analysis are among the younger preserved medieval portals, and represent woodcraft in decay (Gjærder 1952: 41). The county of Telemark belongs to one of the most tradition-bound regions and ornamentation here is increasingly an echo of early medieval art. That a specific symbolic meaning can be dedicated to an ornament once it has become ‘traditional’ art is uncertain. By the beginning of the 13th century the Gothic style was on its way in, but it can be seen from the dates of the portals in this analysis (Table 1) that Telemark kept up the old traditions longer than the central regions of Norway. The quality of the portal ornaments is a problem which is beyond the scope of this paper.

Motifs as Metaphors

To interpret the portal motifs as symbols has been tried by many different researchers. Therefore, I will present an alternative interpretation of the motifs which considers their metaphorical, rather than symbolic, associations. Animal fragmentation and composition of the material culture has earlier been linked with linguistic concepts such as metaphors, whereby words and concepts are rewritten for the sake of poetic effects. Domeij (2004: 150) has looked at the correlation between animal ornamentation and verbal expressions in early medieval Skaldic poetry. She concluded that if the aforementioned animals and body parts, in Skaldic verses where animals occur, were picked out and put together as a whole, a hybrid occurs similar to the compositions of the ornamentation of Scandinavian Iron Age art. In this way, Domeij believes that we can deepen our understanding of animal ornamentation.

On the basis of metaphorical thinking the identified motifs on the portals give some associations. The first association is the ‘supernatural’. Dragons are fabulous beasts of legend and folklore, and are depicted on portals as winged serpents with or without legs. Along with Sigurd on the Lundevall portal, the dragon is depicted as an unnaturally large snake without wings and legs. The fact that Sigurd fought against a dragon is itself a supernatural phenomenon, and may be associated with a fight against forces that may seem insurmountable, since they are not of this world.

The second association is the ‘unnatural’. The animals depicted in the portal ornaments, whether they are real animals or supernatural imaginary animals, have a behavior that is not natural to the respective animals. Large four-legged animals are laying still and pout at the spectator. Dragons and snakes are fighting for their lives. Animals are placed symmetrically towards each other, silently witnessing that they are guardians of the entrance. Men are fighting an impossible battle against giant carnivores.

The third association is the ‘natural’. Plants wind together, horses follow their owners and birds are taking off from the grounds and plains. One element which is difficult to determine, however, is the mask without context.

To look at the motifs as metaphors can help us interpret them. To see motifs as metaphors implies that they really belong in a different frame of reference. The motifs had previously worked at another level of meaning and their expressions must be associated with the reference framework they originated in.

The Ecclesiastical and the Secular

Art is found in many different contexts and rendered in many different media. Previous research concludes that Christian and secular art in medieval Norway are connected, and were not very different. Hauglid (1950: 71) and Gjærder (1952: 15) believe there was no profound distinction between Christian and secular art and architecture, and that the artistic effects that prevailed on the secular doors had a certain connection with the ecclesiastical art and must be seen in relation to this. Karlsson (1976: 41) also assumes that the basic Roman forms and motifs are common to secular and ecclesiastical architecture. My analysis supports these claims, while providing a more nuanced picture. Table 4 presents the results from the analysis of the distribution of motifs on church portals compared with secular portals.

Table 4

Motifs on church portals compared with profane portals

Motifs on stave church portals Motifs on storehouse portals
Snake Snake
Ribbon-shaped animal Ribbon-shaped animal
Four-legged animal Four-legged animal
Plants Plants
Horse Horse
Lion’s head
Battle between dragons and snakes
Biblical motifs
Battle between four-legged animals
Sigurd the Dragon Slayer

Based on Table 4 there is an inequality in the distribution of motifs. The biblical motifs (Figure 5) can be seen as a matter of course in ecclesiastical context and it is not surprising that they are not found in secular context. If they existed, they are not preserved today. Designs that appear only in a secular context are: the battle between four-legged animals, masks and Sigurd the Dragon Slayer. Designs that include Sigurd do appear on stave church portals other than those included in this analysis.

Why some motifs are used in both contexts, while others are reserved for one context only cannot be answered without taking several factors into consideration. Along with the use of motifs, the building’s function indicates the portal’s metaphorical significance. It is likely that the portal distinguishes between more than just inside and outside. That is why there are pictures on the portal. The portal is an important place to express essential elements of iconography. Church portals probably had a stronger position as a metaphor than secular portals. It is only on the church portal that biblical motifs are preserved and the church door in contemporary literature was awarded a symbolic meaning beyond any practical function.

In the Stave Church Sermon it is said that the church door ‘indicates the true faith that guides us in to the common Christian company’. Therefore there is no doubt that the church door, and thus the portal, was a key element in Christian symbolism. It is not just a door leading into the church space but also a door which leads believers into the Christian community. The starting point for the door’s allegorical details goes back to Gregory the Great (Hjelde 1990: 296). First and foremost, the door symbolised faith and a set of three doors symbolised faith, hope and love. The door could also symbolise Christ (John 10: 7–9). The interpretation that the portal symbolises faith is made clear by the Stave Church Sermon.

The biblical scenes in the ecclesiastical context may have had the same function as the ones Andås described in her analysis. The designs were known to medieval people, who were reminded of the Bible stories as they went through the portal. The message could mean that the person was obliged to follow the moral of the story by going through the portal, and to choose salvation rather than condemnation. The same could also apply to the story of Sigurd the Dragon Slayer, who here has been interpreted as a hero, whereby the act of killing the dragon was understood as that the hero fighting and conquering evil. A reminder of Sigurd’s story could involve a call for people to live as good Christians and to constantly fight the evil around them. This appeal has been placed on both the religious and secular portals, and could therefore serve as a general moral - whilst remaining a good story - in all strata of society.

Why the motifs mask and battle between four-legged animals are only on the profane portals and why dragon, bird, lion head and the battle between dragons and snakes are only located in the church context is more difficult to answer. Christensen (1998: 265) states that masks were often carved above the door of a storehouse and loft, and may have been reserved for secular buildings. The dragon, on the other hand, is part of the widely used Sogn Valdres design and this composition may have been reserved for stave church portals. Where medieval people got the models for this composition scheme remains unanswered.

When it comes to the snake, horse, ribbon-shaped animals, four-legged animals and plants, which are found in both contexts, these are classic animal and plant motifs in their simplest form. They are inherited from the Viking styles and are, as expected, found in all medieval art. The horse especially was seen as a powerful animal (see Oma 2000), and the four-legged animals can represent several different species. Both the ecclesiastical and profane portals have four-legged animals as guards on either side of the entrance. These animals are often the only three-dimensional elements on the portals, and face each other symmetrically (Gjærder 1952: 38). Their three-dimensional quality makes them project out from portal, and makes them visible from a distance (see Figure 1). Thus, the viewer’s visual relationship with the animals was established a good while before reaching the portal and they were clearly visible to the person who went through the portal.

The definite reason for applying Sigurd the Dragon Slayer and other motifs on the portals cannot be determined (Nordanskog 2006: 221). Some interpretations are complex and it is questionable whether 12th and 13th century people perceived all the messages we currently read into the pictures. Nordanskog believes that the images could be interpreted as Christian symbols, royal propaganda or stories about heroes from a lost time depending on who was viewing them. Markus (2003: 189) discusses whether the level of understanding of medieval iconography was different for the customer, creator and farmer. She assumes that the public reaction to the art was changing in the 12th century, but that it cannot be compared with today’s human individualistic agency. Can it be concluded that all motifs were understood differently depending on who regarded them in the medieval period? This probably does not apply to the customer, who ordered the design and had a sense of what the motifs symbolized, or what he/she wanted them to be perceived as. I would suggest, on the basis of their ambiguity and the difficulty in interpreting them that the motifs were metaphors meant to give the viewer associations.

We have seen that the portal can legitimately be interpreted as a metaphor. Subjects adapted to the context and the church portal was given their own meaning which differed from other parts of the building. A church portal was also a liminal zone where rituals took place and iconography was viewed.


Several designs from Viking animal styles were reused in the medieval Romanesque style on wooden portals in Norway. The style had changed with the establishment of Christianity but the motifs still existed in their simplest form. Their symbolism is constantly discussed but is understood differently by different people at the time. On the ecclesiastical and secular portals, there was a certain inequality in the distribution of motifs, but generally the selected motifs’ locations were significant to the portal’s liminality and had an impact on the people who entered through it.

From six relatively late portals from the county of Telemark, I identified 18 individual designs. Compared with motifs in the late Viking styles, nine were common motifs in the two periods, eight were exclusively in the Norse animal art, and nine were exclusively on the medieval portals. Generally the animal motifs like the snake, dragon, bird, horse, ribbon-shaped and four-legged animals were used in both periods. This indicates that animal motifs were resistant elements when they were used in their simplest form.

The extent to which motifs are interpreted as symbols is difficult to answer. The interpretation of animal styles emphasizes that animals symbolized the mentality people attributed to them. Research alternates between a symbolic interpretation and a perception that the motifs were only decorative and meaningless. Sigurd the Dragon Slayer has been much debated, but in my opinion he can, when he appears on medieval portals, be understood as a hero. An attempt to understand the motifs on the basis of classical Christian interpretations clarifies that the portal motifs cannot be interpreted in the same way as motifs in Renaissance paintings, in which each scene has an enacted symbolic meaning. To interpret portal motifs likewise underestimates the complexity of the early Christian community in Norway because the portal was more than just art. The portal’s function as an entrance and exit to the building suggests that the motifs had to complement this function.

Comparing the motifs of the ecclesiastical and profane portals reveals that motifs in their simplest form as a snake, four-legged animals, plants and horses are inherited from the Norse animal art and existed in both contexts. It seems that the mask has been reserved for profane portals, while the dragon and biblical motifs were reserved for church portals. The biblical motifs would be recognized by medieval people, who were thus reminded of the moral act in the story. To walk through the portal and enter the church, could commit one to follow the morals and choose salvation rather than condemnation. Although it did not appear in the analysis, Sigurd the Dragon Slayer is a scene that existed in both contexts. Also Sigurd’s history can be seen as a story about moral strength. Sigurd fought against the dragon, winning over evil. In line with the biblical motifs, Sigurd could remind people about fighting for good when they went through the portal.

The use and placement of designs, along with the building’s function, indicates the portals metaphorical significance. The church portal distinguished between outside and inside the house of God, and between the church and the consecrated cemetery. The profane portal in storehouses and lofts distinguished between the storage room, where food and clothes could be stored safely, and the open courtyard. That the portals functioned as a liminal zone, as the threshold between inside and outside, is reflected in contemporary literature. The Stave Church Sermon emphasized that the door was the pivotal element that leads believers into the Christian community, into the church. If the moral of the narrative stories was reinforced by going through the portal, the iconography is meaningful in a liminal perspective.

Designs can tell us about human perception of life in both the Viking and medieval periods. The art’s basic theme in the context of the door itself lay deep in human consciousness.


I would like to express my gratitude to Julie Lund and Lotte Hedeager, my supervisors on this project, and to the Institute of Archaeology, Conservation and History at the University of Oslo where this project was undertaken. I am also greatly indebted to Jack Hunt and Emma Nielsen for their comments on earlier drafts of this paper, though all errors and assumptions remain my own.


1  Andås, M S. (2007). Art and Ritual in the Liminal Zone In:  Andås, M S, Ekroll, Ø; Ø and Haug, A; A, Petersen, N H N H (eds.),   The Medieval Cathedral of Trondheim. Turnhout: Brepols, pp. 47.

2  Anker, P. (1997).  Stavkirkene. Deres Egenart og Historie. Oslo: Cappelen.

3  Anker, P. (1998).  Folkekunst i Norge. Oslo: Cappelen.

4  Berg, A. (1989). Allment Oversyn In:  Norske Minnesmerker. Norske Tømmerhus frå Mellomalderen I. Oslo: Landbruksforlaget.

5  Berg, A. (1991).  Hus for Hus. I Telemark Austre Delen. Norske Minnesmerker. Norske Tømmerhus frå Mellomalderen III. Oslo: Landbruksforlaget.

6  Berg, A. (1993). Hus for Hus. I Telemark Vestre Delen In:  Norske Minnesmerker. Norske Tømmerhus frå Mellomalderen IV. Oslo: Landbruksforlaget.

7  Blindheim, M. (1965).  Norwegian Romanesque Decorative Sculpture 1090–1210. London: Alec Tiranti.

8  Blindheim, M. (1972).  Sigurds Saga i Middelalderens Billedkunst. Oslo: Universitetets Oldsaksamling.

9  Christensen, A L. (1998). Det Middelalderske Boligmønsteret In:  Horgen, J, Skovholt, L; L and Westbye, K K (eds.),   Middelalderen på Romerike. Årbok XIX. Skedsmo: Romerike Historielag.

10  Dietrichson, L. (1892). De Norske Stavkirker. Studier over deres System, Oprindelse og Historiske Udvikling In:  Et Bidrag til Norges Middelalderske Bygningskunsts Historie. Kristiania: Cammenmeyer.

11  Dietrichson, L. (1902).  Omrids af den Kirkelige Munstarkæologi med Særligt Hensyn paa den Norske Kirke. Kristiania: Steen’ske Bogtrykkeri og Forlag.

12  Domeij, M. (2004).  Det Bundna – Djurornamentik och Skaldediktning i Övergången Mellan Förkristen och Kristen Tid.  Gotlänskt Arkiv, : 146.

13  Ekroll, Ø; Stige, M; Havran, J. (2000). Middelalder i Stein In:  Kirker i Norge. Oslo: Arfo.

14  Ferguson, G. (1989).  Signs and Symbols in Christian Art. Oxford: Oxford University Press.

15  Fett, H. (1909). Norges Kirker i Middelalderen In:  Gammel Norsk Kultur i Tekst og Billeder. Oslo: Norsk Folkemuseum.

16  Fredriksen, P D. (2005). Karet/Kroppen/Identiteten In:  Forsberg, L, Larsen, J; J and Østigård, T T (eds.),   UBAS Hovedfag/Master nr. 1. Universitetet i Bergen, Arkeologisk Institutt, pp. 187.

17  Fuglesang, S H. (1980). Some Aspects of the Ringerike Style In:  A Phase of 11th Century Scandinavian Art. Odense: Odense University Press.

18  Fuglesang, S H. (1981).  Stylistic Groups in Late Viking and Early Romanesque Art.  Acta Ad Archaeologiam et Artivm Historiam Pertinenta 1 : 79.

19  Fuglesang, S H. (1982).  Early Viking Art.  Acta Ad Archaeologiam et Artivm Historiam Pertinenta 2, : 125.

20  Gell, A. (1999). The Art of Anthropology. Essays and Diagrams In:  London School of Economics Monographs on Social Anthropology 67. London: The Athlone Press.

21  Gjærder, P. (1952). Norske Pryd-dører fra Middelalderen In:  Universitetet i Bergen skrifter 25. Bergen: A.S. John Griegs Boktrykkeri.

22  Hansen, L E. (1995). Ornamentikken i Overgangen Mellom Vikingtid og Middelalder In:  Institutet för Arkeologi. Uppsala: Uppsala Universitet.

23  Hauglid, R. (1950). Akantus. Fra Hellas til Gudbrandsdal In:  Riksantikvariatet. Oslo: Mittet & Co..

24  Hauglid, R. (1969).  Norske Stavkirker. Oslo: Dreyer.

25  Hauglid, R. (1973).  Norske Stavkirker. Dekor og Utstyr. Oslo: Dreyer.

26  Hauglid, R. (1976).  Norske Stavkirker. Bygningshistorisk Bakgrunn og Utvikling. Oslo: Dreyer.

27  Hedeager, L. (1999). Skandinavisk Dyreornamentik: Symbolsk Repræsentation af en Førkristen Kosmologi In:  Gansum, T, Opedal, A; A and Fuglestvedt, I I (eds.),   Et Hus Med Mange Rom: Vennebok til Bjørn Myhre på 60-årsdagen. AmS-Rapport 11A. Stavanger: Arkeologisk Museum i Stavanger, pp. 219. DOI:

28  Hedeager, L. (2004). Dyr og Andre Mennesker - Mennesker og Andre Dyr. Dyreornamentikkens Transcendentale Realitet In:  Andrén, A, Jennbert, K; K and Raudvere, C C (eds.),   Ordning mot Kaos. Studier av Nordisk Förkristen Kosmologi. Lund: Nordic Academic Press, pp. 219.

29  Hedeager, L. (2011).  Iron Age Myth and Materiality. An Archaeology of Scandinavia AD 400–1000. London: Routledge.

30  Hjelde, O. (1990).  Norsk Preken i Det 12. Århundre: Studier i Gammel Norsk Homiliebok. Oslo.

31  Hohler, E B. (1981). Stavkirkeportalene - Den Dekorative Skurd In:  Berg, K (ed.),   Fra Oseberg til Borgund. Norges Kunsthistorie. Oslo: Gyldendal, 1 pp. 253.

32  Hohler, E B. (1984). Drager i Tre og Sten. Refleksjoner over Stilpåvirkning og Fiksering av Former In:  Vellev, J (ed.),   Romanske Stenarbejder 2. Højbjerg: Hikuin, pp. 119.

33  Hohler, E B. (1995).  Stavkirkeportalene og Deres Symbolikk.  Foreningen til Norske Fortidsminnesmerkers Bevaring Årbok 149 : 169.

34  Hohler, E B. (1999a).  Norwegian Stave Church Sculpture I. Oslo: Scandinavian University Press.

35  Hohler, E B. (1999b).  Norwegian Stave Church Sculpture II. Oslo: Scandinavian University Press.

36  Karlsson, L. (1976). Romansk Träornamentikk i Sverige In:  Acta Universitatis Stockholmiensis 27. Stockholm: Almqvist & Wiksell.

37  Klæsøe, I S. (2002). Hvordan de Blev Til. Vikingetidens Stilgrupper fra Broa til Urnes In:  Nielsen, K H, Vellev, J J (eds.),   Nordeuropæisk Dyrestil 400–1100 e. Kr. Højbjerg: Hikuin, 29 pp. 75.

38  Kristoffersen, S. (2000).  Sverd og Spenne. Dyreornamentikk og Sosial Kontekst. Kristiansand: Norwegian Academic Press.

39  Kristoffersen, S. (2010).  Half Beast - Half Man: Hybrid Figures in Animal Art.  World Arch-aeology  42 (2) : 261. DOI:

40  Liepe, L. (1989).  Sigurdssagan i Bild.  Fornvännen 84, DOI:

41  Markus, K. (2003). Den Mångtydige ‘Majestatis’. Att se på Bilden Genom Medeltidsmänniskans Ögon In:  Danbolt, G, Laugerud, H; H and Liepe, L L (eds.),   Tegn, Symbol og Tolkning. Om Forståelse og Fortolkning av Middelalderens Bilder. København: Museum Tuscalanums Forlag.

42  Nicolaysen, N. (1853–1855).  Mindesmærker af Middelalderens Kunst i Norge. 

43  Nicolaysen, N. (1861). Norske Bygninger fra Fortiden In:  I Tegninger og Med Text. Tredie Hefte. Christiania: Foreningen til Norske Fortidsmindesmerkers Bevaring.

44  Nicolaysen, N. (1903).  Norske Fornlevninger. Christiania: Grøndahl. 2

45  Nielsen, K H; Kristoffersen, S. (2002). Germansk Dyrestil (Salins stil I–III). Et Historisk Perspektiv In:  Nielsen, K H, Vellev, J J (eds.),   Nordeuropæisk Dyrestil 400–1100 e. Kr. Højbjerg: Hikuin, 29 pp. 15.

46  Nordanskog, G. (2006). Sigurd Fafnesbane och Stavkyrkoportalerna: Medeltida Återbruk av en Förkristen Myt In:  Andrén, A, Carelli, P P (eds.),   Odens Oga. Mellan Människor och Makter i det Förkristna Norden. Helsingborg: Dunkers Kulturhus.

47  Nygaard, E M. (1996). De Romanske Steinkirkene i Telemark. Beskrivelse, Analyse og Tolkning In:  Magistergradsavhandling, Institutt for Kunsthistorie og Klassisk Arkeologi. Oslo: Universitetet i Oslo.

48  Ødeby, K. (2012).  Gjennom Portalen - Motiver fra Dyrestilene på Treportaler i Middelalderens Telemark.  Unpublished thesis (MA). University of Oslo, DOI:

49  Oma, K. (2000). Hesten i Nordisk Jernalder. Ei Kontekstuell Analyse av den Symbolske Sfære Kontra den Materielle Røynda In:  IAKK, Det Historisk-filosofiske Fakultet. Oslo: Universitetet i Oslo.

50  Panofsky, E. (1982).  Meaning in the Visual Arts. Chicago: The University of Chicago Press.

51  Salin, B. (1935).  Die Altgermanische Thierornamentik. Typologische Studie über Germanische Metallgegenstände aus dem IV. bis IX. Jahrhundert, nebst einer Studie über irische Ornamentik. Stockholm: Wahlström & Widstrand.

52  Staecker, J. (2006).  Andrén, A, Jennbert, K; K and Raudvere, C C (eds.),   Heroes, Kings and Gods. Discovering Sagas on Gotlandic Picture-stones.  Old Norse Religion in Long-term Perspectives. Origins, Changes and Interactions. An International Conference in Lund. June 3–7, 2004, Lund: Nordic Academic Pres, Sweden pp. 363.

53  Shetelig, H. (1950).  Religionshistoriske Drag fra Vikingetidens Stilhistorie.  Viking XIV, : 49.

54  Wilson, D; Jensen, O K. (1980).  Viking Art. London: George Allen and Unwin.