Christian Medieval Art in Norse Greenland Crosses and Crucifixes and their European Antecedents

During a spectacular excavation in 1921 at the Norse farm of Herjolfsnes (Ikigaat) on the southern tip of Greenland, Poul Norlund found 58 wooden crosses of drift­wood in the graves at the site. Th ...


Introduction
It is well known that the Norse inhabitants of Greenland contributed to the art of the Middle Ages by supplying raw materials in the form of walrus and narwhal ivory for exclusive carving in Europe (Roesdahl 1995and 1998, Frei et al. 2015, Star et al. 2018, Barrett et al. 2019).But the Norse population also developed in artistic terms, albeit at a different, more modest level.A significant number of the wooden crosses, which were mainly found in association with churchyard graves and of which seven bear runic inscriptions, are simple.However, amongst them are four crosses with details, which despite their simplicity, exhibit traits that can be cautiously identified in European prototypes, and the same applies to two crucifixions.The main sections of this article will describe and analyse these six objects and their European antecedents.Dating is a problematic area.How long a delay in the stylistic traits was there, given the considerable distance across the Atlantic and uncertain maritime connections?
In a corrective exercise, we will therefore, where this is possible, also compare the art-historical/typological dating framework with the archaeological dates.A discussion like this, of a combined group of medieval arte facts, has not been undertaken before.Firstly, we will present the con-Nyborg, Ebbe and Jette Arneborg. 2020. Christian Medieval Art in Norse Greenland: Crosses and Crucifixes and their European Antecedents.Scripta Islandica 71: 155-176.
text of the crosses and theories regarding their use, which will be followed by a description and analysis of the individual crosses and crucifixes.

Churches, churchyards, and wooden crosses
The Norse settlements in Greenland were established from Iceland at the end of the 900s and consisted of two settlements on the west coast of the enormous island (figure 1).Accounts in the Icelandic sagas mention Leif Erikson and King Olaf Tryggva son of Norway as initiators of the introduction of Christianity in Green land around the year 1000 (Magnusson & Pálsson 1965: 86).How ever, archaeological investigations of the earliest churches suggest that the Ice landic colonists -or many of them -were already Christian when they came to Greenland, and that they erected quite small church buildings on their farms soon after settling (Arneborg 2012).The small churches were made of stone, turfs and driftwood, the only wood available, and had their closest counterparts in the homeland, Iceland.The Greenlandic bishop's seat at the farm Gardar (today's Igaliku), at the head of the Igaliku Fjord, was estab lished around 1124.According to the account, Greenland's first bishop was appointed by Bishop Asser of Lund in Denmark, but with the establish ment of the archiepiscopal see at Nidaros in Norway in 1152, Green land was incorporated into the Norwegian ecclesiastical province, together with Iceland and the Faroe Islands, and as far as we know, all the bishops who were ordained at Gardar were Norwegians (Arneborg 1991).Writ ten and archaeological evidence indicates that churches must have been erected at least 18 farms at the Norse settlements, and that at least six farms lost their churches during the 1200s.The churches associated with the first settlement of Greenland did not differ from the profane buildings.
It is only the surrounding, circular churchyards that reveal the function of these buildings.Probably at the beginning of the 1100s, and perhaps at the same time as Greenland's incorporation into the Norwegian ecclesiastical province and the establishment of the bishop's seat at Gardar, larger churches were introduced with Romanesque ground plans, consisting of a nave and narrower chancel (two-cell buildings), and from around 1250-1300, proper stone-built, nave churches (one-cell buildings) gained a foot hold, following the development in Europe, especially in Norway.
Nothing is known of the churches' furnishings, such as altars and images.The only archae ological finds associated with the functions of the churches are pieces of broken-up bells.The churchyards are especially interesting to us.In the earliest, the dead were lain side by side.Only in rare cases a grave has cut down into an earlier one, and this may indicate that the graves were marked above ground in some way, although we do not know how.In the later churchyards, in which graves can be very close together, a number of graves were marked by flat tombstones, which covered the whole grave, or by upright stones at the head and foot (for example, Nørlund & Stenberger 1934: 39ff.).Such stone monuments, often with carved crosses, were also obviously associated with European types (figure 2).
Some of the deceased were buried in wooden coffins, whilst others were just wrapped in cloths.Finds of textile remains in nearly all churchyards suggest that the latter was the most common.However, only in the churchyard at Herjolfsnes (Ikigaat), in the southernmost part of the Eastern Settlement, was the preservation so favourable that almost intact  (Nørlund 1924;Østergård 2004).The church at Herjolfsnes was of the Romanesque type and in use until the settlement's depopulation in the middle of the 1400s.The church is by far the most significant finds location for the wooden crosses.No less than 58 crosses, varying in size between 69 ↔ 28.7 cm and 10.5 ↔ 7.5 cm, have been found here.Almost all of these are quite simple, with the cross arm joined at half wood onto the vertical stem of the cross and attached with a wooden rivet or inserted through a crack in the stem.Only three crosses did not have a peg.

The function of the crosses
Poul Nørlund, who was in charge of the archaeological investigations at Herjolfs nes in 1921, interpreted the crosses as "absolution crosses" made for the occasion and placed with the body to ensure the absolution of the deceased (Nørlund 1924: 64).He explained their pegs at the bottom as imitating the pegs on European processional crosses of metal, which were also used as altar crosses.He did not rule out the possibility that some of the large crosses might have functioned as processional or altar crosses before they were placed in the graves.He regarded the crosses as a chrono logical simplification and degeneration of the European crosses.
In a general examination of the Greenlandic crosses and their inscriptions in 1984, Marie Stoklund discussed whether the crosses initially could have been pushed down into the stony soil of the churchyard, and therefore have functioned as grave markers (Stoklund 1984: 104).This might explain the pegs at the bottom of their vertical stems but does not clarify the function and significance of the pegs -the crosses could have been pushed straight down into the ground at the churchyards without the tenon.Neither does this explain how a churchyard cross ended up being placed with its eventual owner, who had just been buried.Stoklund also ends up distancing herself to some extent from the theory.She instead suggests that the crosses were originally used in a "profane" context at the farms.She argues that there is no direct connection between the inscriptions and the dead (Stoklund 1984: 110ff.), and suggests, with reference to the excavations at Sandnes at the Western Settlement in 1932 (Roussell 1936), that crosses were actually also found in a non-ecclesiastical context.Stoklund therefore interpreted the crosses as devotional crosses or "hand crosses" from the homes, which were secondarily placed in the graves, perhaps with the owner of the cross when they died.Here, she sup ported her argument by referring to Norwegian archaeologist Aslak Liestøl, who in 1980 published several similar wooden crosses recovered from archaeological excavations in Bergen.Many of the Bergen crosses also have a tenon for the purposes of attachment, and none are associated with churches or churchyards.Liestøl's interpretation was that the crosses served a function in daily life, and had been used to ward off evil and "placed upon something" at home (Liestøl 1980).The placing of the crosses in graves was therefore secondary, and in a similar way, the extraordinary crucifix group from Sandnes would have secondarily ended up in a grave after having originally been attached to something vertical (figure 10).
Poul Nørlund's view, that these are essentially burial crosses made for deposition with the bodies in connection with the burials, is solidly supported by the find circumstances at Herjolfsnes, whilst his explanation of the pegs can be debated.Stoklund's interpretation, on the other hand, assumes that these might originally have been either churchyard crosses or devotional crosses in homes, and thus presupposes an extensive and less probable secondary deposition of crosses in graves.It is undoubtedly clear from the circumstances of the finds that some of the crosses were used in a "secular" context at the farms.Two finelymade wooden crosses, for instance, were found in the living room at the Farm beneath the Sand at the Western Settlement (one shown in figure 9), and wooden crosses were also found in dwellings at the neighbouring farmstead Sandnes at Kilaarsarfik (Roussell 1936;Berglund 1998).The crosses that are found in a secular context can be explained by the very scattered distribution of churches at the Norse settlements, which meant that there was a long and difficult way to church, especially in winter.People went to church only rarely and needed devotional crosses at home, although this explanation cannot apply at Sandnes, where one of the Western Settlement's two churches was located.So, the crosses would to some extend have been used at the farmsteads and must have been made there.Regarding the more elaborate crosses and especially the crucifixes, which are examined below, we might imagine that these were in some way produced within the estab lished church system, although this is peculiarly partly contradicted by a cross from the Farm beneath the Sand (figure 9) and a crucifix from Austmanna dalen (figure 12), which were found at farm sites located far away from any churches.

The four capital crosses
The unimpressive cross A92 from Herjolfsnes (figure 3) has ends in the form of antiquising, Doric capitals, which consist of an inner, conical part, the echinus and an outer, rectangular abacus.This type of cross, referred to here as a capital cross, can also have several inserted profiled parts.It is known from AngloSaxon book minia tures, such as the socalled Arenberg Passion (Raw 1990, Pl. II), and the well-known manuscript depiction dating to after 1031 of King Canute the Great and his Queen AElfgyfu/Emma, who are seen donating an altar cross to the New Minster at Winchester (figure 4).A number of exquisite Ger man capital crosses and crucifixes of gold with precious stones also date to the same time, the earliest of which is thought to be a magnificent piece known as the Lothar Cross and dating to c. 1000, in the treasury of Aachen Cathedral (figure 5) (Grimme 1973: 24-28).
There is obviously no direct connection between such exclusive crosses and Greenland.As a European phenomenon, the capital ends date to the 1000s.But they are also known slightly later in Scandinavia, where the golden crucifix from Lisbjerg in Denmark may be dated around 1100 (Nør lund 1926: 61-72, in which the Greenlandic crosses are also referred  to).Capital ends are, however, primarily known from wooden sculpture in Nor way.The earliest is probably a crucifix from Leikanger Church (figure 6), which is also dated to c. 1100, and the latest example is perhaps the socalled "Ottonian" cross (figure 7), which the expert Martin Blindheim has dated to the second half of the 1100s (Blindheim 1998: 28-29, 34).As it is a cross without a Christ figure, it can be directly visually compared with the Greenlandic examples, even though its capitals are slightly more com plex in their profiles.Christian Medieval Art in Norse Greenland Journeys to Greenland mostly started from Bergen in Norway (Magerøy 1993), and particularly after the Greenlandic bishop's seat came under the juris diction of Nidaros, it would seem likely that the cultural contacts would have been very one-sidedly Norwegian, and that the capital motif came from here.Given the obvious distribution in Norway, it would certainly be quite unlikely if a Greenlandic woodcarver had come up with the design independently.
Another Greenlandic cross from P. Nørlund's excavations at Herjolfsnes in 1921 (figure 8), which is made of two crosses, has identical cross ends in the form of Doric capitals.It bears runic inscriptions, both on the vertical stem and the horizontal arm; one text is in Norse language and the other is appar ently Latin.On the stem is the owner's name "Brakil" (Brigit?) and an invocation of Saint Michael and the Virgin Mary, whilst on the cross arm is a magical formula, which is upside down.According to runologists, the inscrip tions can be attributed to two different rune carvers, but the stem and cross arm are only slightly different and are therefore probably parts of two very uniform crosses (Stoklund 1984: 110-111;Imer 2017: 226-227).The stem is smooth, with a profile inserted under the echinus, which has concave sides.The cross arm, on the other hand, has contoured lines along the edge, and the capitals are slightly larger, with a narrower profile.The similar character of the capitals suggests that the two original crosses may have been made by the same woodcarver.Their joining together can certainly be interpreted as a repair, using two strikingly uni-form crosses.Lastly, there is a capital cross from the Farm beneath the Sand at the Western Settlement (figure 9).Its elongated capitals are a vari a tion of those on the cross with the inscriptions, D10664 (figure 8), with a concave echinus, and the sides of the cross are unusually carved convex (Berg lund 1998).
The two crosses from Herjolfsnes (A92 and D10664, figures 3 and 8) were both found in the churchyard.The firstmentioned cross, A92, is an unstrati fied stray find from 1849, whilst the cross with inscriptions, D10664, was found in the churchyard west of the church (Nørlund 1924;Jóns son 1924: 280ff.).The cross from the Farm beneath the Sand is interesting because of its "secular" context.It was found in 1993 during the excavation of the farmstead's living room (room 5), which dates to the 1300s.As the settlement was abandoned at the end of this century, the cross cannot be later in date.
It cannot be ruled out that an AngloSaxon cross or crucifix reached Green land directly from Great Britain and constituted this type, but all the evidence suggests that the four Greenlandic capital crosses can be attri buted to cultural influence from Norway.The crosses can apparently be typo logically traced right back to the first generations of settlement on Green land.Bearing in mind the dating of the Norwegian objects and a possible additional delay in Greenland, a date within the 1100-1200s -and after the establishment of the Greenlandic diocese -might be estimated.The archaeological dating of the cross from the Farm beneath the Sand, on the other hand, is placed in the 1300s, with a possibility of a further long "lifespan".The Latin text on the cross with the inscriptions (D10664) also points in the same direction (Imer 2017: 228).This would indi cate a dating to around 1300 or even later.The two crucifixions A remarkable crucifixion group from Sandnes at the Western Settlement is carved from quite thin spruce, measures 34.5 ↔ 13.3 cm (0.5-1.8 cm thick) and has originally been fastened onto something vertical with two holes (Roussell 1936: 171) (figure 10).This is therefore not a traditional church yard cross.
The fine and detailed piece in openwork -unfortunately damaged in several places -is surrounded by foliage in an almond shape, which tapers slightly downwards.Jesus is hanging from the cross with slightly sloping arms and his head leaning to the right.He seems to be wearing a crown of thorns.The eyes are shut under frowned eyebrows, the ribs are marked, a turbulently folded loincloth has baggy overhangs at each hip.The position of the feet cannot be ascertained.The sides of the cross are contoured and the ends, certainly above and below, have a sort of leafed capital, from which large sprouts of acanthus emanate.At the bottom are flanking side figures, Mary and John the Apostle, who stand in traditional mourning poses.Mary is wearing a headcloth and leaning towards her son.Her right hand leads towards him and the left up to her cheek, like on a Norwegian crucifix group from the manor Austråt by Trond heim (Blindheim 2004: 56-57).Only small bits of John the Apostle are preserved.The naturalistic depiction of the persons, the hanging position and agonised expression of Jesus, as well as the penchant for acanthus leaves, point towards the transition between the two earliest phases of Gothic (Early and High Gothic), which in Europe occurred in the decades around the middle of the 13th century.The crown of thorns, if present, became common in the 1230s in Europe, although not until after around c. 1250 in Scandinavia.Crucifix figures with small overhangs at the hips appear in a short period around 1250, and there are related crucifix figures in Musèe Schott in Brussels and in Kiaby Church, Scania (Didier 1982: 150-151;Liepe 1995: 30, 293).When we examine the details, the overall depiction seems to be mostly associated with a number of Norwegian crucifixes such as one from Tretten in eastern Norway from 1275-1300 (Blindheim 2004: 37-39, 190-91) (figure 11).
This work again seems to have depended on English book miniatures dating to around the middle of the 1200s exemplified by the Ames bury Psalter.Here, we can identify basic characteristics that resemble those on the Greenlandic crucifix, although the figure from Tretten hangs with completely outstretched arms and its feet are turned outward in pain.The ends of the Norwegian crucifix are halved quatrefoils that terminate in a Doric capital, a trait which is not found on its own at this time.Halved quatre foils combined with this type of capital is probably a specifically Nordic form, which is known from Denmark and is even more common in Norway (Blindheim 2004: 120, 130, 150 and 165).There are therefore once again reasons for us to assume that a Norwegian influence is present.The traits we have observed are so chronologically specific to the middle and second half of the 1200s, that we can perhaps imagine that the Greenlandic piece was made around 1300.We can rule out that the object is an import.
Another Greenlandic crucifix that was found at farm V53d at the Western Settlement (Roussell 1941: 269), is in low relief on the background of a panel, from the ends of which the plain cross slightly protrudes (figure 12).This can be categorised as a panel crucifix.In the cross centre behind the head is a halo with curved incisions, which must indicate a cross.The figure is hanging low with completely straight arms, the head is upright, and its right foot is laid over the outstretched left with its side facing forwards.Under a narrow crown of thorns, long hair with diagonal incisions surrounds a summarily carved face with small eyes (open or closed?) and a straight nose and mouth.The nipples are slightly hanging, and the loincloth has a roll around the waist and a curved overhang at the front.It forms cornets on both sides with omega hems.As the figure, despite its lowhanging arms, appears upright, this is an almost Romanesque trait dating to before 1200, and the same applies to the marking of the nipples.The unusual position involving the side of a foot facing forwards can be traced back to French crucifixes dating to the beginning of the 1200s such as in Le Mans and Nevers (Schneider-Berren berg 1977: 43-65).The figures of these are, however, depicted stand ing, with slightly separate feet.With other, slightly later figures that hang in the Gothic way, the position is maintained, although the right foot is placed on top of the left and the nail is struck though both feet like on the figures in St Katharinenthal, Switzerland, and at Brindisi, Italy (Schnei derBerrenberg 1977: fig.101, 113).It is this design that the cruci fix from Aust manna dalen is associated with.Nordic parallels consist of a Nor wegian example from Skaun (Blindheim 1998: 80) and two Danish cruci fixes in Rybjerg and Lomborg in North Jutland (Danmarks Kirker Ring købing Amt 2001-2010: 749-750).The early Gothic way of draping the loincloth is also characteristic of the mid-13th century and the crown of thorns (as a Nordic phenomenon) dates to after this.Another late trait is the way the figure hangs low, as can be seen in an illustration in Villard de Honne court's well-known sketchbook from c. 1230 (Scheller 1995: 176-187), but in Scandinavia dates to the 1300s.If the idea of using a panel for the background of the crucifix is not local to Greenland, it should be mentioned that crucifixes are also found in Europe where stems are expanded to form a side panel.The phenomenon is primarily Italian, although the examples found in the North include several Danish ones (Dan marks Kirker Ribe Amt 1988-91: 276-279;1994-2003: 3013-15).Our crucifix is an unusually instructive example of stylistic delay within an isolated environment.It should obviously be dated from its latest traits and therefore most likely can be placed in the 1300s; the cultural influence is once again apparently Scandinavian, and probably Norwegian.
The two crucifixes from the Western Settlement can also only be loosely archaeo logically dated.The group cannot be archaeologically dated at all, as it is an unstratified stray find from the churchyard at Sandnes (Kilaarsarfik).The panel crucifix was found in a "profane" context in a farm complex at the head of Austmannadalen (farm V53d).It was in the fill of a bench in a room dating to the farm's last phase of use, just before the settlement was abandoned at the end of the 1300s (Roussell 1941: 247).Its late dating therefore corresponds well with the art-historical evidence.
Apart from pieces of broken church bells, a crosier of walrus tusk, and a golden ring in a bishop's grave in the church yard of the cathedral no imported liturgical accessories have been found in the Norse Greenland.Even though situated far way from the church centres in Norway we might presuppose that at least the cathedral possessed a few holy crucifixes and processional crosses that inspired local artists and non-professional carvers, and styles may have wandered from farm to farm over long periods of time.Some of the carvers might never have seen the original pieces.

Conclusion
The attempt to challenge the art-historical/typological dates with the archaeo logical ones has only been partially successful.The crosses from Herjolfs nes and the crucifixion group from Sandnes were all found in church yards without definite stratigraphic information.The cross from the Farm beneath the Sand is typologically dated to the 1100-1200s but was found in a dwelling room dating to the 1300s.It may already have been old at this point.But much would speak to the fact that the capital cross would have had its roots far back in time in Greenland.Lastly, the only chronological correspondence applies to the crucifix from Aust manna dalen, whose art-historical dating to the 1300s is supported by its find location in a living room dating to just before the abandonment of the Western Settlement.
The Greenlandic crosses and crucifixes reflect a craft environment that was extremely locally based and not particularly specialised.The material, washed ashore wood and timbers, was however not just carved out of basic crafts man ship, but in the case of the crucifixions also with a certain degree of expertise and art.The delay in style testifies to an environment that was very isolated due to the considerable distance to Europe.It is, Christian Medieval Art in Norse Greenland how ever, clear that quite specific impulses were received from across the Atlantic, and the mixture of stylistic traits, especially on the panel crucifix, demonstrate that the impulses could be "stored" over time, perhaps utilising models in the form of paintings and sculptures of different dates in the church interiors of the settlements.The crucifixes and images of saints in the churches may, like in Iceland, have been a mixture of imports and domestically made objects (Blindheim 1997).If in Greenland people had thus begun to produce their own images for the churches, local woodcarvers may have directly or indirectly contributed to the design of the small crosses and crucifixes.A further perhaps rather bold explanation might be that artistic Greenlanders were familiar with model illustrations, such as those in the Icelandic illustrated manuscript from the Late Middle Ages (Fett 1910;Scheller 1995: 241-249;Kristjánsdóttir 2013).Here, we find models or prototypes for Biblical scenes and saints in the "modern" stylistic designs of the 1400s, mixed with long outdated forms dating to as far back as the High Middle Ages.
The investigation has cast new light upon a society that was the furthest out post of the Christian world.The crosses and crucifixes that we have analysed represent the boundary for the influence of medieval art, and more than this.A further chronological and geographical dimension can be added.Important traits of the crosses and crucifixes were medieval inher i tance from the classical, antique world: Traits, such as the Doric capital, the acanthus, and the naturalism in rendering of bodies and garment folds.The Norse settlements "at the end of the world" would thus represent the very earliest occurrence of antique art and culture in the Americas.

Fig. 1 .
Fig. 1.The Norse settlements on the west coast of Greenland.Furthest to the south was the Eastern Settlement, whilst the Western Settlement was located near today's Nuuk.Base map from Nunniffiit, Greenland National Museum & Archives' register over preserved ancient monuments.

Fig. 4 .
Fig. 4. King Canute the Great of Denmark and England, with his wife AElfgyfu/ Emma, donating a cross to the altar in New Minster, Winchester.Miniature in its Liber Vitae, 1031.© British Library Board.

Fig. 8 .
Fig. 8. Cross with inscription from Herjolfsnes.Made from driftwood.Height: 25 cm.Inventory number D10664.The cross was transferred from the National Museum in Copenhagen to the Greenland National Museum & Archives in Nuuk in 2001.Photo: Michael Nielsen, Greenland National Museum and Archives

Fig. 9 .
Fig. 9. Cross found at the Farm beneath the Sand, dating to the second half of the 1300s.Made from driftwood.Height: 14 cm.Inventory number KNK1950x1208.Photo Michael Nielsen, Greenland National Museum & Archives.

Fig. 10 .
Fig. 10.Crucifix group (calvary) found in the churchyard at Sandnes in 1932.Made of spruce, Height: 34 cm.Photo: The National Museum of Denmark.

Fig. 12 .
Fig. 12. Panel crucifix from Austmannadalen.Made from driftwood.Height: 21 cm.In the National Museum, Copenhagen.Inventory number D12812.Photo: The National Museum of Denmark.