Religious Images and Iconoclasm in Reformation Iceland

: This work assesses what happened to liturgical objects from Icelandic churches and monastic houses during and after the Lutheran Reformation, through an examination of written sources, such as inventories and Visitation books, and material evidence in museum collections and from archaeological excavations. The aim of this work is ﬁrst, to assess the extent and nature of iconoclasm in Iceland and secondly to re-examine traditional narratives of the Icelandic Reformation in the light of material culture.


Introduction
Judging solely from the vast collection of visual heritage from the Catholic period preserved in Iceland, it would seem that the pillaging and demolition of religious imagery that followed the Protestant Reformation in many European countries did not transpire there. Nevertheless, there is evidence to suggest that some attacks on images and liturgical objects took place during and after the Reformation. It also seems that in some cases, objects were transformed and reused in ways that better suited the Lutheran faith and in yet other cases, objects of precious metals were stolen and shipped to Denmark as taxation for the king.
As elsewhere in Europe, church interiors and inventories in Iceland changed significantly in accordance with Lutheran conventions. The material culture of the churches and the changes thereof present valuable evidence in illuminating the nature of the changes that followed the Protestant Reformation and the pace at which these changes took place. As argued by Jürgensen, textual sources might provide us with "the reasoning and vocabulary of the reformers", but churches and their material culture reveal the ways in which reformers "expressed their ideas when it came to actually giving form or body to their visions of a new church free of the 'popery' they reacted against" (Jürgensen 2017(Jürgensen , pp. 1041.
This work assesses what happened to religious images and devotional objects during and after the Lutheran Reformation in Iceland through an examination of written sources, such as church inventories and Visitation books, and material evidence in museum collections and from archaeological excavations. The aim of this work is first, to assess the extent and nature of iconoclasm in Iceland; was iconoclasm confined to a few isolated and erratic attacks or did a systematic confiscation and destruction of Catholic objects take place following the Reformation? What were the motives and targets of iconoclasts in Iceland? How were objects otherwise re-contextualized? Secondly, this study aims to re-examine traditional narratives of the Icelandic Reformation in the light of material culture; what can the evidence at hand tell us about the process of the Reformation in Iceland and its political implications? Was the Icelandic Reformation first and foremost a political struggle, centered around the monarch's acquisition of power? Furthermore, perhaps most importantly, what role did objects play in the advancement and resistance to Lutheranism? To gain a deeper insight into the political aspect of the Reformation, a wider Scandinavian context will be also considered.

The Reformation in Iceland and Denmark-Norway
On the morning of the 7th of November in 1550, Jón Arason, the last Catholic bishop of Hólar, and two of his sons, Ari and Björn, were beheaded in Skálholt, the other episcopal see of Iceland ( Figure 1). A year later, in 1551, the diocese of Hólar submitted to the Lutheran Church Ordinance of King Christian III 1 which had been accepted in Skálholt in 1541. Consequently, the king became the head of the church, and church property was largely secularised (Ísleifsdóttir 2013, pp. 353-56). The execution of the three men marked a symbolic concluding event in a more than decade-long political struggle between the Catholic clergy in Iceland and Protestant Reformers working in the interest of the king (Hugason 2018, p. 173;Ísleifsdóttir 2013). With the death of Jón Arason "the external opposition to the Reformation was broken" and Lutheranism could be firmly established in the country (Andersen 1990, p. 156). Amongst the first advocators of the Reformation in Iceland was Gissur Einarsson, who had become acquainted with Lutheranism during his studies in Hamburg in the period 1531-1534. Two years after his return home he was appointed assistant to Ögmundur Pálsson Bishop of Skálholt but, in the interim, remained silent about this Protestant sympathies. Shortly after, in 1539, when Ögmundur, due to his old age and impaired eyesight, resigned his position, he recommended Gissur as his successor and in 1540 Gissur was appointed as the bishop of Skálholt (Guttormsson 2000, p. 54).
An important factor that differentiates the Reformation in Iceland from mainland Europe is that there was no evangelical movement prior to the formal conversion and thus it has been argued that, for Icelanders, the Reformation initially presented itself as a political struggle. Because of the low population density and absence of cities and market towns, a middling class of merchants and craftsmen, that often were the first to take to the message of Protestant Reformers in mainland Europe, was practically non-existent in Iceland, and there were no universities. Thus, it has been argued that there was little sociocultural premise for any Evangelical movements and, therefore, Iceland was "completely Amongst the first advocators of the Reformation in Iceland was Gissur Einarsson, who had become acquainted with Lutheranism during his studies in Hamburg in the period 1531-1534. Two years after his return home he was appointed assistant to Ögmundur Pálsson Bishop of Skálholt but, in the interim, remained silent about this Protestant sympathies. Shortly after, in 1539, when Ögmundur, due to his old age and impaired eyesight, resigned his position, he recommended Gissur as his successor and in 1540 Gissur was appointed as the bishop of Skálholt (Guttormsson 2000, p. 54).
An important factor that differentiates the Reformation in Iceland from mainland Europe is that there was no evangelical movement prior to the formal conversion and thus it has been argued that, for Icelanders, the Reformation initially presented itself as a political struggle. Because of the low population density and absence of cities and market towns, a middling class of merchants and craftsmen, that often were the first to take to the message of Protestant Reformers in mainland Europe, was practically non-existent in Iceland, and there were no universities. Thus, it has been argued that there was little sociocultural premise for any Evangelical movements and, therefore, Iceland was "completely unprepared for the Reformation" (Andersen 1990, p. 154). However, around the time Gissur Einarsson came to power, some Lutheran influence had gained footing in Skálholt and in 1540 the first Icelandic translation of the New Testament by Oddur Gottskálksson was printed (Guttormsson 2000, pp. 51-52).
In 1538, shortly after Christian III had proclaimed the new Church Ordinance for Denmark, he attempted to introduce it in Iceland, but it was rejected by both Ögmundur Pálsson, Bishop of Hólar, and Jón Arason, Bishop of Skálholt (Andersen 1990, p. 155). In 1541 a royal emissary was sent to Iceland to have the ordinance accepted, in which it succeeded only in Skálholt which Gissur Einarsson was now in possession of. Ögmundur, who had turned against the new bishop, was arrested by the emissary and died on the journey to Copenhagen (Guttormsson 2000, pp. 57-59). In the meantime, some dramatic measures had been employed to put an end to Catholicism in the country and on the morning of Pentecost in 1539, Sheriff Didrik von Mynden and some other men working for King Christian III are said to have stormed into the monastery of Viðey, plundered anything of monetary value and struck and burned the rest (Diplomatarium Islandicum (DI) 1857-1972, X, pp. 478-80). During construction work near the monastic site in recent years, an effigy of St. Dorothy was found by chance. The head of the effigy is missing but other than that the effigy is whole (Gunnarsdóttir and Kristjánsdóttir 2016, p. 36). The headless effigy of St. Dorothy may thus present a tangible remnant of the attack on Viðey, led by Didrik von Mynden, which marked the dissolution of the monastery (See Kristjánsdóttir 2017, p. 326).
Bishop Gissur Einarsson died in 1548, only 36 years of age. In his short time in office, he had made great efforts to organise the diocese of Skálholt in accordance with the new church ordinance. He sternly opposed any Catholic ceremonies and, amongst other things, urged priests to obtain the Icelandic translation of the New Testament from 1540. In addition, Gissur had evangelical sermons translated into Icelandic, and himself translated parts of the Old Testament (Andersen 1990, pp. 155-56). After the death of Gissur, Jón Arason attempted to take possession of the diocese of Skálholt to re-establish Catholicism and in doing so, took prisoner Gissur's successor. Soon after that, Jón and his sons were arrested by the king's men and executed (Ibid.).
The execution of Bishop Jón and his sons has traditionally been regarded as the resolution of Roman Catholicism in Iceland and is commonly described as a turning point in Icelandic religious and political history. However, as various scholars have pointed out in recent years, the Reformation was a long and complex process, encompassing several realms of society, and did not straightforwardly end with the acceptance of the new church ordinance in 1551 but continued to develop over the next decades, or even centuries (Ísleifsdóttir 2013;Hugason 2018). In its narrowest sense, it has been suggested that the Reformation, strictly seen as the establishment of the Lutheran Church, took place in the decades between 1540 and 1600, from the first attempts of Christian III to introduce the new Church Ordinance at Alþingi until Lutheranism had become somewhat established in both dioceses (Guttormsson 2000, p. 110).
Another result of the low population density in Iceland was that the changes that followed the Reformation happened at a slow pace. This was also due to the fact that Iceland was, at the time, far removed from the sovereignty of the king and there was no central authority capable of effectively taking matters into their own hands and leading the change. With the exception of the envoys sent to Iceland in 1541 and 1551, the presence of the king was limited and did by no means suffice to propel forward the Reformation at the speed which the traditional narrative seems to suggest (Hugason 2018, p. 183).
It can be expected that the changes following the Reformation happened at a different pace in different localities depending on the willingness of parishes to adapt to the new tradition, and although the bishops were required to make regular visitations to each parish, the management of the churches was first and foremost in the hands of the priests. This is clearly demonstrated by an account of Sigurður Jónsson, the youngest son of Jón Arason, who served as a priest at Grenjaðarstaðir, in the diocese of Hólar, who shortly before the Easter of 1554 officially stated that he would begin serving in accordance with the Lutheran doctrine instead of the Catholic one. It seems that Sigurður was amongst the first Icelandic priests who had been ordained by the Catholic Church but willingly converted to Lutheranism (Hugason 2015). According to a register of priests from the period it seems that only about a quarter of priests in the diocese of Skálholt and a fifth of the priests in the diocese of Hólar resigned their position as a result of the Reformation, which means that a majority of the priests serving in the new Lutheran Church had previously served as Catholic priests (Hugason 2018, pp. 187-89). Therefore, it seems likely that it generally took some years and perhaps a new generation of priests for any significant changes to take place (Ibid.).
The situation in Iceland was in many ways comparable to that in Norway where, following the victory of the Reformation in Denmark, Lutheranism was introduced in 1537. It has been maintained that the transition in Norway, for the most part, took place without much force from the Danish king who recognised the need for a slow and gradual change (Von Achen 2020, pp. 80-81). However, whereas the Reformation in the Danish-Norwegian kingdom may have been a political success for the king, it has been argued that Lutheranism did not gain a strong footing amongst the populace. Indeed, the Reformation in Norway has been described as a 'Reformation without people' (Laugerud 2018;Von Achen 2020, p. 82).
The progress of the Reformation in Denmark and Norway will not be traced in detail here. It suffices to say that a systematic destruction of images and devotional objects from the Catholic period did not transpire on a large scale following the Reformation and there seems to have been a moderate tolerance towards images within the kingdom, although a few iconoclastic riots broke out in Copenhagen, Malmö, and Schleswig in Denmark in the years leading up to the Reformation and in Bergen in Norway a few decades later (Johannsen and Johannsen 2012, p. 266; Von Achen 2020, pp. 80-81; Gilje 2011; Figure 2). 2 However, although such attacks were indeed very rare, the description of the Danish Carmelite Paul Helgesen (b. ca. 1485) of the attack on the Church of Our Lady in Copenhagen in 1531 suggests that they may have been dramatic (Quoted in Frederiksen 1987, p. 117. Own translation): First they threw over all the Holy images, spat on them, blew them with their fists and mocked them with ghastly insults, while they smashed them with axes, thereafter they permeated the choir where they demolished completely the Canon's chair and the panel. [ . . . ]. They even went so far, as to rip the books apart.
The series of iconoclastic conflicts which took place in Bergen in the period 1568-1572 were part of an attempt to introduce a second Reformation under Calvinistic influence. During these conflicts, the altars in many churches in Bergen were stripped of images of saints under the lead of Bishop Jens Skielderup (b. 1510-d. 1582) (Gilje 2011). Skielderup went so far in removing what remained of the Catholic tradition that the town council feared that he would "strip the churches down to the mere walls" (Ibid., p. 74. Own translation). Importantly, the removal of images from the churches of Bergen was largely initiated by the Danish nobility, specifically by the king's commission that, during an inspection of the town, criticised Skielderup for not having already taken steps to have them removed. As a result of the pressure posed on him by the king's commission, Skieldrup called together a synod where he decreed that images and effigies were to be removed from churches, albeit with some few exceptions (Ibid.). Similar to Bergen, influence of the Danish monarch affected the tolerance towards images in the province of Jämtland which had ecclesiastically belonged to Sweden but was transferred to the Norwegian diocese of Trondheim in 1571. Consequent to the transferral, in the late 16th and early 17th century, many churches in Jämtland received new altarpieces and other church fittings which were decorated with catechisms in Danish instead of images of saints. It seems that in the Danish-Norwegian Jämtland there was generally less tolerance towards images than in the provinces east of it which remained Swedish, and while in the Swedish providences the old and the new were commonly mixed together, the priests in Jämtland more often disposed of Catholic imagery upon obtaining the new pieces. It has been noted that the removal of religious images from the churches of Jämtland was initiated by the authorities without popular support (Holm 2017, pp. 389-90). This is demonstrated by an incident in the parish of Offerdal in the north of Jämtland, where images of saints were disposed of by the vicar but salvaged by some members of the parish who continued their worship in non-official, semi-domestic chapels (Ibid; Zachrisson 2019, pp. 8-11). Thus, although the events in Bergen and the province of Jämtland are perhaps not representative of the general situation in Norway, they demonstrate the national politics at stake at the Reformation within the Oldenburg Monarchy.

Iconoclasm: Motives and Approaches
Shortly before his death, Gissur Einarsson, the first Lutheran bishop in Iceland, issued an edict which was to be read in the churches in the diocese of Skálholt and included instructions for the confiscation of all effigies worshipped by "ignorant and supersticious" Similar to Bergen, influence of the Danish monarch affected the tolerance towards images in the province of Jämtland which had ecclesiastically belonged to Sweden but was transferred to the Norwegian diocese of Trondheim in 1571. Consequent to the transferral, in the late 16th and early 17th century, many churches in Jämtland received new altarpieces and other church fittings which were decorated with catechisms in Danish instead of images of saints. It seems that in the Danish-Norwegian Jämtland there was generally less tolerance towards images than in the provinces east of it which remained Swedish, and while in the Swedish providences the old and the new were commonly mixed together, the priests in Jämtland more often disposed of Catholic imagery upon obtaining the new pieces. It has been noted that the removal of religious images from the churches of Jämtland was initiated by the authorities without popular support (Holm 2017, pp. 389-90). This is demonstrated by an incident in the parish of Offerdal in the north of Jämtland, where images of saints were disposed of by the vicar but salvaged by some members of the parish who continued their worship in non-official, semi-domestic chapels (Ibid; Zachrisson 2019, pp. 8-11). Thus, although the events in Bergen and the province of Jämtland are perhaps not representative of the general situation in Norway, they demonstrate the national politics at stake at the Reformation within the Oldenburg Monarchy.

Iconoclasm: Motives and Approaches
Shortly before his death, Gissur Einarsson, the first Lutheran bishop in Iceland, issued an edict which was to be read in the churches in the diocese of Skálholt and included instructions for the confiscation of all effigies worshipped by "ignorant and supersticious" people (Biskupaannálar Jóns Egilssonar 1856, pp. 87-88). An exception was made for effigies depicting Jesus, Virgin Mary and the Apostles, which could be kept as tokens of remembrance. However, Gissur quickly took a less tolerant stance and a year later he journeyed to the church of Kaldaðarnes where he had the Holy Cross taken down. The cross, which was believed to be miraculous and had long been an attraction for pilgrims from around the country, was stored in Kaldaðarnes for a few years, until Bishop Gísli Jónsson (b. 1515-d. 1587) had it moved to Skálholt where it was broken into pieces and burned (Ibid.). The episode that supposedly followed was described by the poet Bjarni Jónsson (b. 1560-d. 1640) (Bragi Óðfraeðivefur n.d., Own translation): All writings and ornaments Torn apart and burned; Christ's images shattered, Paper and icons rotten.
Much has been written about the iconoclastic destruction of images and sacred objects during and after the Protestant Reformation in Europe and the motives that drove such attacks (e.g., Aston 1988Aston , 1989Duffy 1992;Eire 1986;Koerner 2002). Iconoclasm varied much in intensity between regions and whereas in some places, both statuary art and two-dimensional images from the Catholic period have largely remained intact, in other places a significant part of the visual heritage from that time has been destroyed completely or damaged greatly (Graves 2008, p. 35). The latter was the case in England where the reformation was accompanied by a wave of plundering and destruction encouraged by the authorities. This wave reached its crest during the reign of Edward VI, when a systematic destruction of religious images was implemented throughout the whole country with the lead of Archbishop Cranmer and his associates (Duffy 1992, pp. 448-77). In some cases, objects were shattered or burned so that nothing remained of them. In other cases, images were defaced or painted over, and the heads, hands and feet of statues were hewn off, their eyes gouged out or noses scraped off. Many images subjected to such attacks were deliberately left to view in these states and can, to this day, be found in many parish churches and cathedrals around England (Ibid., p. 35).
The critical attitude of many Lutheran theologians and reformers towards images in the first decades after the reformation was first and foremost directed at the veneration of things, which had characterised Christian devotion from at least the 12th century (Eire 1986, pp. 13-14). In this cult of images, paintings, reliefs, relics, effigies, crucifixes and other liturgical objects in churches were regarded as embodiments of divine powers. As such, they were not conceived or treated as inert things but as animated and alive, and capable of conveying power and healing (Bynum 2011, pp. 21-25). The belief in the powers and vitality of sacred objects was invigorated by stories of such objects miraculously bleeding, weeping or changing colours, which proliferated in the late Medieval period (Ibid., pp. 21-22). Often, objects were not only thought to invoke external powers but were in themselves powerful; as in the case of the Holy Cross from Kaldaðarnes, people made pilgrimages to view them and touched them and kissed "as if Christ himself were present in them all" (Ibid., p. 127). After the Holy Cross had been burned, written accounts describe how the old people at Skálholt sought to get ahold of its ashes, demonstrating the power invested in the material traces of the Cross (Biskupaannálar Jóns Egilssonar 1856, pp. 87-88). Another such powerful item in Iceland was the reliquary of St. Þorlákur in the Cathedral of Skálholt, which people stroked with their hands before placing them on their eyes or other parts on their body that required healing. Regularly, the reliquary was carried around the cemetery, followed by people singing canticles or reading scriptures (Ibid.).
The Catholic cult of images was criticised by many Protestant reformers who feared that the veneration of objects misdirected people's worship to the material world, to the point that it overshadowed the worship of God (Eire 1986, p. 54). In the view of Protestants, objects could not "objectify a spiritual reality" and thus, worship offered to images "offered only to the artistic representation, not to the person represented by it" (Ibid., p. 59) The aim of iconoclasts was, at least in part, to demonstrate this and to reveal the powerless and inert nature of images (Graves 2008, p. 39). Protestant reformers were, in other words, not only opposed to the veneration of saints, which were often the subject of religious imagery, but to the worship of material things and attempts to objectify the divine. However, Luther himself did not wish to ban images from churches altogether as, for example, Karlstadt and Zwingli did; he considered images to be neither good nor bad and if necessary, they could still be used within a certain framework as long as they were not worshipped. It is of interest whether this is reflected in the Icelandic material and whether iconoclasts exclusively targeted objects which were seen as particularly sacred or miraculous, such as the Holy Cross at Kaldaðarnes.
As already discussed, iconoclasm did not take place on a large scale in Denmark. Efforts were focused on modifying unsuitable objects rather than demolishing them completely; church furniture from the Catholic period certainly found its place within the Lutheran faith, and Bishop Peter Palladius (b. 1503Palladius (b. -d. 1560) recommended that altar tables were used as foundations for pulpits and that altar cloths were used to make clothing or bandages for the poor and sick (Johannsen and Johannsen 2012, p. 267). One interesting example of modification of imagery is an altar piece made in 1496 which originally belonged to the monastic church of Esrum, in the north of Zealand but was moved to the church of St. Olai, in Helsingør in the east of Zealand in 1659. In the mid-17th century, an Abbot Peter which is depicted on the bottom of the piece was transformed into a Lutheran priest; the tonsure was covered up, a beard and a moustache were added to his face, his robe was changed and a ruff was put around his neck (Jensen 1921, p. 182;Skinnebach 2016, pp. 157-58). This suggests that it was not always the images in themselves which were considered unsuitable but rather their subject. However, a critical attitude towards images and the veneration of objects gained some footing in Denmark and in reforming the Danish church art, images were often replaced by text, as recommended by Luther (Johannsen and Johannsen 2012, pp. 262-65). As discussed earlier in the example from Jämtland, words became a popular decorative feature in Danish churches, particularly from in the late 16th century and onwards, and quotes from scriptures, often in gilded letters, were displayed on a great number of church walls, and altar pieces and fronts ( Jürgensen 2017( Jürgensen , pp. 1055. As expressed by Niels Palladius (b. 1510-d. 1560), brother of the bishop (Quoted in Jürgensen 2017, p. 1055): The true ornament of the Christian churches is the sacred office itself of God's words, and the administration of the sacraments and pious ceremonies to whom they are joined, not logs, a wooden nose or eye, painted and fashioned faces.
In some cases, the addition of text to unsuitable imagery could even offer a compromise. In 1593, an alabaster retable in Vejrum Church in Jutland depicting St. Catherine was, for example, 'neutralised' with the addition of the words: 'Saints, you should not serve, but worship God alone. These images are embellishment alone. They have no other power, nor virtue' (Ibid., p. 267). With the addition of the text, attempts were made to reveal the inert nature of the image and to strip it of its power; it could still be enjoyed as a piece of art or even as a historical curiosity (see Jensen 1921, p. 196), but in no instance was it to be worshipped or bestowed with divine powers.
As indicated by the account of Gissur Einarsson, some opposition towards the worship of objects seems to have been present in Iceland. However, scholars do not agree upon the extent to which this opposition materialised through iconoclastic attacks or confiscation of devotional objects; while some have regarded the story of the crucifix of Kaldaðarnes as a testimony of systematic destruction of church art and religious objects (e.g., Björnsson 1964) others reject the idea of such vandalism altogether (e.g., Kristjánsson 2017). Only a few studies have looked at what happened to devotional objects during and after the Lutheran Reformation in Iceland. The most influential of these is a study by medievalist Cormack (2017) where she examines church inventories of two Lutheran bishops. As will be discussed further, Cormack finds in these inventories some indications of vandalism as well as other important evidence for the fate of Catholic objects in the decades after the Ref-ormation. However, Cormack does not look in depth at any extant objects from the Catholic period. Theologian Gunnar Kristjánsson (2017) has also examined selected inventories of churches as well as some Icelandic church art in the national museums of Iceland and Denmark but claims to find no examples of vandalism. Gunnar is highly critical of accounts about iconoclasm in Iceland and believes them to be untrustworthy dramatizations of the true events. However, Gunnar examines no other factors than iconoclasm and does not consider how the objects may otherwise have been recontextualised. Apart from these two, a few other authors have examined the change in church interiors and art after the Reformation (e.g., Harðardóttir 2017), but they have mostly been concerned with how the Lutheran faith influenced new church art and furnishings rather than the fate of the Catholic objects.

The Fate of Catholic Objects in Iceland
In examining the fate of religious images, statues and other devotional objects, church inventories and visitation books provide important evidence. In the aforementioned research on possible iconoclasm in Reformation Iceland, Margaret Cormack (2017) uses data from the visitation books of two Lutheran bishops of Skálholt, those of Gísli Jónsson (1559-1587) and Brynjólfur Sveinsson (1639-1674). In the inventories of Brynjólfur, Cormack finds several mentions of broken effigies; in the inventory of the church of Keldur, he, for example, noted "three heads broken off alabaster" and "three fragments so the images can not be identified". (Ibid., p. 247). In the same church, Brynjólfur listed a broken effigy of John the Apostle (Ibid., p. 250). Above the altar in the church of Þykkvibaer Brynjólfur found "remains from alabaster effigies" and in Mýrar there were four effigies, one of which was broken. At Hvol, Brynjólfur also mentions a "broken stone altarpiece" which may have been decorated with images of saints (Ibid.).
Interestingly, the damaged objects listed by Brynjólfur seem to have been only partly damaged and then retained within the churches in that state rather than demolished completely. Only a few effigies of saints from the Catholic period which are preserved in the National Museum of Iceland show signs of such vandalism. Amongst those is a majestic statue depicting the sedes sapientiae missing the Christ child. The statue is carved in oak and is 90 cm high. It has been dated to the 13th century and is believed to originate from Sweden. In the right side of head of the effigy is a large cleft and one of the lilies is missing from the crown (Sarpur n.d., 10944/1930-355). Art historian Selma Jónsdóttir, who studied the statue, argued that this damage was purposefully inflicted with a blunt object (Jónsdóttir 1964, p. 9). Upon examination, this indeed seems probable. The statue was gifted to the National Museum of Denmark in 1853 but returned to Iceland, along with many other artefacts in 1930. According to Danish accounts, the National Museum of Denmark acquired the effigy from a Gudmann Jr. who came by it in Hjaltadalur, in which the Cathedral of Hólar was situated. In the accounts of the National Museum of Iceland from 1930 this is, however, claimed to be a misconception but no further explanation to this claim is provided (Sarpur n.d., 10944/1930-355). Selma Jónsdóttir questioned this and argued that statue was indeed imported to Hjaltadalur, in the episcopacy of Guðmundur góði Arason (Gudmund the Good, b. 1203-1237) who is said to have been devoted to Virgin Mary. Selma maintains that the inhabitants of Hjaltadalur regarded the statue as particularly sacred and that its special status eventually drove someone to attack its face with an axe or a similar tool (Jónsdóttir 1964, pp. 56-57). It is difficult to confirm the association of the statue to Hjaltadalur or to Guðmundur góði but it does indeed seem likely that it was attacked due to its special status.
Other than this effigy, very few of the religious images exhibited in the National Museum seem to have been subjected to iconoclastic attacks. Many statues of saints are missing hands, arms, or parts of their face, such as the nose, which were common targets of iconoclasts, but this is more likely due to the fact these parts are most liable to break off, rather than deliberate destruction. Nevertheless, such damage, albeit accidental, reflects a changed attitude towards the images which were, in the Catholic period, regarded as sacred but were neglected and left to decay in the centuries after the Reformation.
Only two of the numerous effigies recorded in Sarpur, an online database for Icelandic museum collections, show possible signs of deliberate destruction. One of these is an effigy of Paul the apostle carved in oak, which belonged to the church in Hjarðarholt in Borgarfjörður. The effigy is missing its nose and its eyes have been gouged out, which is typical for iconoclasm. Another effigy recorded in Sarpur which might also have been subjected to vandalism is carved in pine and depicts Andrew the apostle. The effigy, which has been dated to the 13th century and was probably made in Iceland or imported from Norway, belonged to the church at Teigur in Fljótsdalur, a turf church dedicated to St. Andrew and the Virgin Mary which was abolished in the late 19th century (Sarpur n.d., 2441(Sarpur n.d., /1883. At some point the bottom half of the effigy was cut off. This is not typical for iconoclasm and thus possibly requires an alternative explanation. It is noteworthy that both these effigies depict apostles, which along with Marian images and images of Christ, were acceptable according to bishop Gissur Einarsson, as long as they were not worshipped. Thus, if these effigies were truly subjected to iconoclasm, it seems likely that they were attacked on grounds of their special status or supposed miraculous nature rather than their imagery, as in the case of the statue associated with Hofstaðir discussed above.
The effigy of St. Andrew is mentioned in an inventory of the church of Teigar from 1397 along with an effigy of the Virgin, a crucifix and an altar stone containing relics (DI IV, p. 78). In an inventory of 1553 made by Bishop Marteinn Einarsson, none of these items are listed (DI XII, p. 650). However, the National Museum of Iceland obtained the effigy from the church of Teigar in 1883 (Sarpur n.d., 2441(Sarpur n.d., /1883; therefore, it seems likely that it was retained in the church throughout the years, perhaps hidden or stored away. If that is the case, it can be inferred that the inventories are not always accurate in that they might not include all the objects kept within the churches, and that either the bishops responsible for the inventories intentionally avoided listing such items or they were hidden away by the parish priest before their visitations. As in the case of the effigy of St. Andrew, a great part of the religious imagery and devotional objects from the Catholic period is missing from inventories shortly after the Reformation. For this study, two collections of inventories were examined in order to compare the listings before and after the Reformation: first, the abovementioned inventory book made by Bishop Marteinn Einarsson during his tour of the diocese in the period 1553-1554, only little over a decade after the Lutheran church ordinance was officially brought into force there. A transcript of the original manuscript of these inventories has been published in the series Diplomatarium Islandicum (DI) (1857( -1972. The second inventory book was written in 1601 by Bjarni Marteinsson, scribe at Skálholt, and contains a collection of inventories of churches in the diocese, the oldest dating to the 12th century and the most recent to the year 1600. The manuscript is preserved in the National Archives of Iceland but no transcript of it has yet been published (Biskupsskjalasafn AII n.d.).
The collection of inventories written by Bjarni is somewhat problematic as only a few of the individual inventories within it include a date and thus, it can be difficult to assess their precise age. However, the collection should not be dismissed completely. On grounds of their contents, the pre-Reformation inventories in the front of the manuscript can easily be distinguished from the 16th century inventories, which suffices for the aim of this study; although the exact dates of the inventories are unknown, they can still be used in comparison to inventories from after the Reformation (see Table 1). The inventories in the collection of Bjarni reveal that in the centuries before the Reformation, the churches in the diocese of Skálholt were filled with effigies of saints, crucifixes and reliquaries. Moreover, in a majority of the inventories, the dedications of the churches to the various saints are mentioned. Already in the inventory from 1553 to 1554 this had changed radically. While many of the church inventories examined from that year include chalices, patens and corporals, only one included effigies, that is the church of Oddi which was said to own an effigy of the Virgin Mary and some other "graven images" (DI XII, pp. 652-53). The inventory from Oddi also includes a reliquary and four crucifixes (Ibid.). The only other mention of a crucifix was in the inventory of Kaldaðarnes which was described as broken (Ibid.,. In addition, the inventory of Skarð lists a reliquary with silver (Ibid., p. 656) and the church at Ásar had hagiographies of St. John and St. Paul (Ibid.). Three altar pieces, which might have included images of saints, were mentioned in the inventories: one in Oddi (Ibid., pp. 652-53), one in Húsafell (Ibid., pp. 666-67) and one in Gufudalur (Ibid., pp. 667-68). Only one inventory notes the church's dedication, that is the church at Keldur, which was dedicated to Paul the Apostle (Ibid., p. 654).
Evidently, a large part of the Catholic objects of the churches in the diocese of Skálholt are missing from the inventories from 1553 to 1554. As described earlier, Bishop Gissur Einarsson did order the confiscation of effigies and other objects worshipped by the people in the diocese. Written sources also cite that in the diocese of Hólar, Ólafur Hjaltason, the first Lutheran bishop, had all effigies and crucifixes which he considered to be objects of worship to be collected and broken (Cormack 2017, p. 243;Skarðsannáll 1922, p. 134). While these narratives have often been seen as unreliable exaggerations of the true events (see Kristjánsson 2017, pp. 218-19;Harðardóttir 2017, p. 195;Hugason 1991;Hugason 1988), the difference in the inventories before and after the Reformation might, at first glance, suggest that there is some truth to them. Importantly, however, as demonstrated in Cormack's study, some objects seem to reappear in later inventories, suggesting that they were omitted from destruction.
Indeed, returning to Cormack's study, one of the most interesting results is the difference in the two inventories she examines: whereas Brynjólfur, writing his inventories in the mid-17th century, lists various effigies of saints and other devotional objects of a Catholic nature, Gísli, who served within the first decades after the Reformation, very similar to Marteinn Einarsson, does not mention effigies, and lists only a few crucifixes and relics as well as two hagiographies (Cormack 2017, pp. 249-50). As has already been postulated, this might suggest that the first Lutheran bishops simply avoided noting such objects in their inventories although they were present in the churches. However, if this was the case, the question remains why Marteinn Einarsson decided to note only a few selected items but circumvented such objects in all the other churches. Another explanation for the disappearance of objects from the inventories is that they were hidden away in the first decades after the Reformation but were slowly reintroduced as Lutheranism had become more established.
It is well known from other countries that devotional objects were hidden away during the Reformation. In England, for instance, a large number of alabaster panels from the Catholic period have been found hidden in roofs, walls or under floors of churches and even in domestic buildings (Cheetham 2005, pp. 53-54). The same seems to have occurred in Denmark; in Odense Cathedral in Funen, for example, a reliquary shrine of St. Canute was hidden in the choir walls and rediscovered in 1582 (Johannsen and Johannsen 2012, p. 266), and in the Cathedral of Roskilde in Zealand some epitaphs were hung over images of the Virgin Mary and some other saints which had been painted on pillars within the church (Lauring 1963).
It is well possible that this was also the case in Iceland. Some decades ago, the farmer of Klaustur by Lagarfljót, near the monastic site of Skriða in the diocese of Skálholt, found an effigy of the Virgin Mary, which has been dated to the Medieval period, in the walls of an old shed. It is not unlikely that it belonged to the monastery of Skriða and was hidden within the wall of the shed to protect it from vandalism or theft or for the fear that it would be confiscated by the church authorities (Sarpur n.d., 14414/1950-145). A 16th century effigy of Christ, removed from the crucifix, found under the floor tiles of the church of Staður in Steingrímsfjörður in the diocese of Skálholt (Sarpur n.d., 1834(Sarpur n.d., -1892, might similarly have been deposited to protect it from harm. At the same time, the deposition of the effigy may have been a part of a ritual to protect the church building from disasters, or even from riots. Similar deposits of devotional objects have been detected in Norway; for example, a textual amulet containing prayers to St. Dorothy and the Holy Cross was recovered from under the floor tiles of a medieval wooden stave church in Torpo. The amulet has been discussed by Hagen, who has speculated that it might have been deposited to protect the church from fires due to its connection with St. Dorothy (Haug Hagen 2019).
In any case, the effigy of the Virgin Mary in the wall at Klaustur and that of Christ under the floor tiles in Staður certainly imply that there was some motive to place objects out of sight, and that priests or parish members may have feared that the belongings of their churches would be confiscated or attacked by iconoclasts or the king's men. A broken effigy of St. Barbara which was found during archaeological excavations at Skriða makes compelling the argument that the effigy of the Virgin was concealed on a nearby farm to safeguard it from a similar fate. All the pieces of the effigy of St. Barbara were found in the choir of the monastic church in 2006, excluding the face which was found on its own near the kitchen in 2007. The fact that the face of the effigy was found separately, in a distinct building, suggests that it was deliberately hewn off, perhaps around the dissolution of the monastery at the Reformation (Kristjánsdóttir 2012, pp. 97-100). In an inventory of the church of Skriða from 1641 an effigy of "some holy maid" was listed, suggesting that the faceless effigy was retained within the church, although it had become unrecognizable to the writer of the inventory (Ibid., p. 100). The effigy is made of terra cotta and was probably produced in Utrecht in the 15th century. The individual pieces have now been attached back together and the effigy stands around 30 cm tall (Ibid.,.
For whatever reason that many of the devotional objects missing from the earliest inventories after the Reformation reappear in the inventories of Bishop Brynjólfur Sveinsson in the mid-17th century, this seems to suggest that the critical attitude towards religious imagery diminished or at least became more moderate some decades after the Reformation.
The same pattern has been suggested in Denmark, where the Protestant material critique diminished as the Reformation lost its potency and the exuberance and grandeur of the Baroque found its way into Danish churches (Jürgensen 2017(Jürgensen , pp. 1052(Jürgensen , 1074. In addition to evidence for broken and damaged images and statues, there is also evidence to suggest that some liturgical objects from the Catholic period were modified as to better fit Lutheran customs. Such physical transformations reveal clearly the cultural transformations that the objects underwent during the Reformation and how their meaning and value changed (e.g., Dooijes and Nieuwenhuyse 2007). An example of a physically modified object is a 14th century manuscript which was originally a Catholic mass-book containing text in Latin. At some point after the Reformation, the Latin text was scraped off the pages of the manuscript and replaced with music notes and Icelandic song lyrics. The original illustrations of the manuscripts were kept in place and merged with the new notes and lyrics. On some pages, attempts were made to fit the new text with the subject of the illustrations but in other cases this was not possible, and the text and illustration do not relate to each other (Ingólfsson 2019, pp. 68-69). Through this transformation, the songbook bore physical marks of its history, and for its readers the illustrations would have conjured powerful memories of the original meaning of the book and the association to other objects, people and events from which that meaning emerged.
Other examples of repurposing of objects from the Catholic period include a piscina from the church of Saurbaer which is said to have been used as a baptismal font after the Reformation, and parts of a monstrance from Skálholt which were recycled as feet for two candlesticks (Harðardóttir 2017, p. 196). In the church of Villingaholt, a case originally used to house an effigy was used for storage (Cormack 2017, p. 250). One remarkable instance of the repurposing of a liturgical object after the Reformation is described in an inventory of the Cathedral of Skáholt from 1698 where some old procession staffs were said to serve as hinges for a door (Harðardóttir 2017, p. 196). All these instances of repurposing reflect a radical recontextualization of the objects, which in the Catholic period were considered sacred but after the Reformation were translated into seemingly mundane objects. However, the translation of these objects certainly created the possibility of keeping them within the churches.
A relic case from the church of Valþjófsstaðir, exhibited in the National Museum, from which large parts of the gold overlay, which originally depicted a crucifixion scene and other religious imagery, has been removed (Sarpur n.d., 3612/1891-91), may present yet another example of modification. Church inventories after the Reformation in fact commonly mention reliquaries which are missing parts; the lid of the reliquary in Oddi noted by Brynjólfur Sveinsson was, for instance, lost and in Kálfafellsstaðir one side of a reliquary was missing (Cormack 2017, p. 248). However, although it is well possible that parts of the reliquaries were removed because of religious motives, they may also have been stolen. It is known that from the 16th century and onwards, religious objects of precious metal from Icelandic churches were shipped to Denmark as taxation for the king and it has been suggested that after the Reformation, church objects were often stolen from churches for this purpose (Kristjánsson 2017, pp. 236-37;Magnússon 2013, p. 34). In either case, it is significant that what remained of the cases was kept in the churches, whether they served as tokens of memory for the old religion or were actively used in religious activities.

Results
In tracing the fate of religious imagery and devotional objects in Iceland, at least three phases can be identified. In the Catholic period they were received as embodiments of divine power and played an important role in exalting the religious experience. Importantly, the significance of these objects stemmed from a certain understanding of their matter as potent and alive. During and around the Reformation the meaning of these objects changed drastically as Protestant reformers sought to reveal the inert nature of what they regarded as mere pieces of wood or stone. With the orientation towards the Word of God, as advocated by Lutheranism, there was less room for the worship of such dead things. In Iceland, this message was received by some of the first Lutheran bishops who were critical, if not hostile, towards the worship of objects and ordered the confiscation and destruction of objects which were perceived as miraculous. However, the bishops do not seem to have been successful in this and there is not much evidence to suggest that systematic attacks on religious objects took place in Iceland.
Nevertheless, albeit neither common nor systematic, some sporadic attacks indeed seem to have transpired. It has been suggested that these exclusively singled out objects which were perceived as particularly sacred or miraculous (Cormack 2017), in accordance with the orders of the bishops and Luther's criticism of the cult of images. However, this is not entirely convincing since many objects which were indeed seen as particularly holy did survive the Reformation (e.g., Eldjárn 1958). It seems more likely that in the few instances that churches were attacked, iconoclasts struck anything they could find. However, it is noteworthy that according to the material evidence and some written sources, the heads or faces of effigies seem to have been a particular target, as, for example, was the case in England, and thus some aspect of iconoclasm was selective rather than erratic. Written sources also suggest that, in some instances, objects were burned or otherwise demolished, but this is impossible to confirm with material evidence.
As a result of the bishops' decrees, and some few iconoclastic attacks, many objects seem to have been either hidden away or physically modified and thereby incorporated into Lutheran customs. In many cases, these were reintroduced into the churches after the havoc of the Reformation. The fact that objects could successfully be hidden or stored away reflects what has been argued about the Icelandic Reformation, that it was not driven by a powerful, central authority but rather contingent on the readiness, or indeed reluctancy, of each parish to adapt to the new religion. In this phase, devotional objects went from being, first and foremost, devotional objects to being laden with political significance.
The narratives from Bergen and in Jämtland in Norway reveal the political implications of religious imagery during the Protestant Reformation. In both cases the stripping of the altars was set in motion by Danish influence and may have played a role in the establishment of Protestantism, and thus the authority of the Danish monarchy. This, in many ways, echoes the situation in Iceland where the Reformation, in some part, revolved around the king's acquisition of power and such efforts manifested themselves, for example, in the violent attack on the monastery in Viðey led by the king's men. The continued worship of images in illicit chapels or the depositing of religious imagery inside church walls or under floor boards, "out of sight, yet not out of mind to the still faithful defenders of the old confession", as expressed by Johannsen and Johannsen (2012, p. 266), may have been powerful acts of resistance, not only towards the new religious conventions but also towards external powers. Whereas the confiscation or destruction of sacred objects could serve to establish the new religion in the favour of the Danish king, hiding objects, or continuing their worship, could serve as a counteract to conserve the old, and signify resistance to Protestantism or to the authority of the king. Thus, although it is hard to say whether the Icelandic people received the Reformation first and foremost as a political struggle, religious objects certainly played an important role in that aspect.
After the first decades of the Reformation, the opposition towards images and the cult of saints seems to have diminished and the objects once again found their place within churches, either as an integral part of religious activities or as physical memories of the Catholic period that had survived even the turmoil of the Reformation. This underlines the fact that Lutheranism continued to develop and change long after the last Catholic bishop was executed and a new church ordinance was accepted, and that the Reformation cannot be understood merely in terms of specific dates or events but rather as a long and complex process. Thus, this examination of Catholic objects in many ways confirms what has already been stated about the Icelandic Reformation: namely, that it developed over a long period of time and advanced in different ways in each parish. Further work is needed to examine in more detail the advancement of the Reformation on local scales. In particular, a detailed study of the parishes in the diocese of Hólar, where the opposition towards the Reformation is believed to have been stronger, is needed as well an examination of changes in devotion in domestic contexts.

Funding:
The research received partial funding by the University of Iceland Research Fund.