2010 B. Gjerland and C. Keller 161 Introduction The literature of the medieval Icelanders gives a unique insight into the history, culture, and religion of the North Atlantic Norse. It conveys the medieval Icelanders’ perception of themselves and their past. For good and bad, it is the insiders’ view (Hreinsson 1997, Kristjánsson 1988, Ólason 1998). Archaeology does not have the same explanatory power as the written word when it comes to exploring the mentality of people in the past. However, despite its shortcomings, archaeology may still convey valuable information about the mindset of the Viking and medieval Norse. Can the Norse peoples’ use of the landscape tell us something about their values and beliefs? Was there really a common religious Norse geography in pagan times, and did it continue into the early Christian era? Or, were there differences in the organization of the religious elements in the landscape between West Norway, Iceland, and Norse Greenland? Although the issues are familiar, these exact questions have not been addressed through landscape archaeology, and no obvious methods for comparisons of landscapes of this kind are available. This was a pilot study which examined and compared the locations of pagan burial sites and medieval church sites in three countries (Fig. 1), based on a limited set of samples from each. The results were then compared to look for similarities and differences. Graves and Churches of the Norse in the North Atlantic: A Pilot Study Berit Gjerland1 and Christian Keller2,* Abstract - Was there a sacred Norse geography in the North Atlantic region during the Viking and Early Middle Ages? In this study, the locations of Late Iron Age pagan Norse graves in West Norway and Iceland are analyzed and compared. The analysis also includes medieval church-sites in these regions, as well as in Norse Greenland. The approach is that of landscape archaeology, and two sets of analyses are used: an attribute analysis where the locations of pagan burials and medieval churches are situated relative to landscape attributes, and a distance analysis where the distances from graves and churches to the original farm cores are analyzed. The results show distinct differences in landscape organization between Norway and Iceland, both concerning pagan burials and church locations. No pagan Norse graves are yet known in Greenland, but the church locations are compared to Norway and Iceland. The church locations in Greenland have features in common with both Iceland and Norway. The political organization and church politics in the three countries are discussed, and the results of the analyses are tentatively associated with historic events. Special Volume 2:161–177 Norse Greenland: Selected Papers from the Hvalsey Conference 2008 Journal of the North Atlantic 1Sogn og Fjordane County Council, Cultural Department. PO Box 173, N-6801 Førde, Norway.2IKOS, University of Oslo, PO Box 1010 Blindern N-0315 Oslo, Norway. *Corresponding author - christian.keller@ikos.uio.no. 2010 Figure 1. The three regions compared in this study: The Eastern Settlement in Norse Greenland (left), Sunnfjord in West Norway (right), and Skagafjörður and Mývatn in Iceland (bottom). 162 Journal of the North Atlantic Special Volume 2 The primary goal was to find methods for comparing landscape locations between the three countries. The second goal is to identify significant differences in the organization of religious elements in the landscape, and to discuss and possibly explain them in relation to the religious and political development of the three countries. The farm was the hub around which the social, economic, and spiritual life in the Norse world revolved. In search of Norse cultic landscapes, pagan or Christian, it is natural to focus on the location of the farms, farm cores, pagan graves, and churches. The sites in the three countries were selected from the Sunnfjord region in Sogn og Fjordane in West Norway, the Mývatn and Skagafjörður regions in Northern Iceland, and the so-called Eastern Settlement and Western Settlement in South West Greenland (Fig. 1). All the sites in West Norway and Iceland were visited in the field. Most of the Greenland sites have been visited earlier; all were analyzed on the basis of published archaeological maps (in Krogh 1982 and in Guldager et al. 2002). Two methods were used to make the comparisons: an attribute analysis where the sites (graves and churches) were classified according to their proximity to certain landscape variables or attributes, and a distance analysis where the distances between the sites and the farm cores were measured. The analysis shows that in West Norway there was continuity in the use of the landscape from pagan to Christian times. In Iceland, the pagan burials are located according to different principles than the burials in West Norway, and there appears to be a break in continuity from pagan to Christian times. No pagan burials have yet been found in Greenland. The early churches in Iceland lie completely integrated among the farm houses (see Krogh 1983). This pattern seems to be the case also for some of the West Norwegian churches. Other Norwegian churches, however, are located in what appears to be more publicly accessible areas away from the farm houses. These church sites differ dramatically from the Icelandic church sites, and are believed to reflect a new principle for church building in the Norwegian Middle Ages. Greenland’s church locations appear to be a combination of the Icelandic and the Norwegian location types, showing affinity both to Icelandic and Norwegian church-building traditions. Visual Landscape Analysis The field-work behind this study was initially based on the principles of visual landscape archaeological methods as they were presented in 1997 (Gansum et al 1997). The methods launched back then were modifications of a landscape architectural approach (as in Lynch 1992), but adapted to serve as landscape archaeological tools. The principle of a visual landscape analysis is to describe how the various elements in the landscape are located relative to each other (what may be seen from where, which sites are visually dominant, and which seem more isolated or remote). In an archaeological context, it is also important to know what constructions were contemporary and what function they had, as well as the relative age of the various installations in the landscape. For this paper, only a selection of the data collected during the field-work was used. This paper focuses on spatial organization and layout, and especially the proximity between various features in the landscape, and also how various topographical elements have been used. Farms The Norse population lived at farms. Farms are places, and most events recorded in the medieval written records are associated with specific farms. The farm core1, i.e., the cluster of houses in each farm unit, was the principal social arena in Norse society. In the present analysis, the ancient farm cores have been tentatively reconstructed using archaeological and/or retrospective historical methods. The historic farm cores sometimes had almost village-like properties; often they also consisted of several holdings2. The West-Norwegian “klyngetun” (clustered farm / historical farm) was perhaps the closest to a village in appearance (e.g., Berg 1968:87–210). Here, sites of the historical farm core are supposed to go back to at least the Middle Ages, perhaps even further. Archaeologically, farm cores from the Early Iron Age have frequently been discovered on present-day farm lands. Farm cores from the Late Iron Age are, on the other hand, rarely found and therefore are assumed to lie in the same place as the historical farm core. Recent archaeological investigations in West Norway have also confirmed this interpretation, both directly and indirectly. In fact the investigated farm cores go back to at least Pre-Roman times (Dokset 2007, Olsen 2010, Øye 2002:69). Towards the end of the nineteenth century, a land reform was launched in Norway to assemble the many small lots of land into larger, more viable units, and new, single-farm cores where built. Detailed farm maps describing the situation before the land-reform and specifying the location of the old farm lots, farm cores, boundaries, home fields, roads, and landing places were used as important source material for this analysis. Viking Period burials in Trøndelag3 may be associated with prehistoric farms (Sognnes 1988:15), but not all the prehistoric farms have archaeologi2010 B. Gjerland and C. Keller 163 cally visible graves. In other regions, the farm-grave relationships are not that simple; many farms have more than one Iron Age cemetery, and others have none. Studies indicate that Late Iron Age graves may represent independent (i.e., inheritable) farms (Iversen 2004:66–71). A farm without graves may, on the other hand, indicate a tenant farm. It should be remembered that most people in the pre-Christian era did not get an archaeologically visible burial. In the attribute analysis in Table 1 and Figure 2, the reconstructions of farm cores contemporary with the Viking Age graves in West Norway were based on the land-reform maps from the end of the nineteenth century. In Iceland and Greenland, the contemporary farm cores were identified archaeologically. The home fields were identified either on the basis of land-reform maps (Norway), or on home-field boundary walls (Iceland), or on the improved fields next to the farm cores (Greenland). Pagan Burials Burial customs in Viking Scandinavia were extremely diverse (Gräslund and Müller-Wille 1993). In Norway alone, some 6000 graves from the Viking Period are known (Solberg 2000:222), some with horses and boats (Müller-Wille 1970). Both cremation and inhumation burials were used in Scandinavia, but cremations decline towards the south and west; they are less common in Denmark, rare in the Norse cemeteries in the British Isles, and nonexistent in Iceland (Graham-Campbell and Batey 1998:144; but see Friðriksson in Eldjárn 2000:594, Richards 2003:391). Graves in Norway are normally marked by a mound or cairn, usually in an easily visible spot— located near the farm core or within sight of it, by the farm road, or sometimes by the boat-house or -landing (Solberg 2000:222). There are also Viking Period graves under flat ground, often with a modest Figure 2. Radar diagram showing landscape attributes connected to burial sites in West Norway and Iceland. For comparison, burials from West Norway dated to Early Iron Age and the General Iron Age are indicated. Eight attributes are indicated clockwise around the diagram; they are not mutually exclusive, i.e., one burial site may feature one attribute, another two or three attributes. Table 1. Landscape attributes connected to burial sites in West Norway and Iceland. For comparison, burials from West Norway dated to Early Iron Age and the General Iron Age are included. Attributes Area Farm core Home field Route Post-box Boundary Water Beach Welcome Raw data Early + General Iron Age in West Norway 8 11 8 3 1 10 8 5 Late Iron Age in West Norway 8 8 0 1 0 5 5 7 Iceland 0 0 8 3 6 7 7 7 Normalized data Early + General Iron Age in West Norway 14.8 20.4 14.8 5.6 1.9 18.5 14.8 9.3 Late Iron Age in West Norway 23.5 23.5 0.0 2.9 0.0 14.7 14.7 20.6 Iceland 0.0 0.0 21.1 7.9 15.8 18.4 18.4 18.4 164 Journal of the North Atlantic Special Volume 2 stone paving. In terms of the grave furniture, the roughly 350 graves discovered in Iceland may be associated with the lower to medium range of Norwegian graves in terms of status. The burials in Iceland are often located by a track, but often also near a boundary, or facing a lake, river, or the sea. Most of them lie under flat ground with little or no preserved surface marking; many have horses and some have boats (Friðriksson in Eldjárn 2000:592–593, 610; Friðriksson 2004). They are notoriously difficult to find. The Icelandic graves may have much in common with the North Norwegian graves from the Merovingian period (560/570–800 A.D.; see Solberg 2000:182–184, 186–188). These are usually inhumation graves and often under flat ground, and the frequency of horses in the burials increases into the Viking Period. Norse Greenland was, according to the written records, settled around A.D. 985. The Saga of Eirik the Red (Eiríks saga rauða) claims Eiríkr was a pagan on arrival in Greenland, and credits his son Leifr with the introduction of Christianity. The credibility of the story has been questioned (e.g., Krogh 1975). No unequivocal pagan Norse burials have yet been found in Greenland, but considering how difficult it is to find pagan graves in Iceland, the question of pagan burials in Greenland must remain open. The burial customs of the Norse changed dramatically with the introduction of Christianity in the tenth to eleventh centuries; all Christians were entitled to burial in consecrated ground. In Scandinavia, Iceland, and Greenland, the dead were inhumed in cemeteries which, like in the British Isles, were established around the churches. Most medieval churches in the Northern world were built at farms, and most of the churches and their parishes still carry their original farm names (Olsen 1928:234–255). Pagan Burials: Attribute Analysis Ten burial sites in Sunnfjord4 in West Norway were selected for comparison with burial sites in Iceland. Previous analyses of these and other sites in the Sunnfjord region were performed by Gjerland in Sigurðsson et al. 2005. The burials were dated to: 1. Early Iron Age (ca. 500 B.C.–A.D. 560/570), 2. Late Iron Age (A.D. 560/570–1050), and 3. Iron Age General (500 B.C.–A.D. 1050; these are burials that belong to either the Early or Late Iron Age, but either the artifacts did not allow a more specific dating, or the cairns have not been archaeologically examined). The attributes selected are (clockwise from top of Fig. 2): Farm core = location among or nearby the houses on a farm; Home field = location at the home field;Route = location at or addressing a major route or track, i.e., between farms or settlements; Post-box = location at or addressing the track from a major road to the farm core; Boundary = location at or addressing a farm boundary on land; Water = location overlooking or addressing either the sea, a lake, or a river; Beach = location at a beach (will often address water); and Welcome = location at or addressing the track leading from a boat-landing5 to the farm core. For comparison between West Norway and Iceland, only graves from the Late Iron Age are relevant, but in order to discuss continuity within West Norway, graves from the earlier periods were included in the analysis. The burials are single, or in small grave fields. Most of the graves do not exist in the landscape today, but the material and locations are recorded in the archaeological archives. From Iceland, burial sites from ten farms in the Skagafjörður and Mývatn regions were selected for visual analysis. Site locations and archive information were kindly provided by Guðný Zoëga, Adolf Friðriksson, and Árni Einarsson. The number of analyzed localities is modest, and the results should therefore be treated with caution. Based on our general archaeological knowledge, however, it is likely that the attributes will prove to be significant also in larger regional studies. The actual material information is presented as raw data and is given in the tables. The presentation in the radar diagrams show normalized data and thus give a relative value of the different attributes compared to each other within each area. The normalized data must, however, be used with caution due to the limited number of analyzed objects. The West Norwegian graves in this material are located close to the farm cores, and they are also on the home fields. This relationship seems consistent throughout the Iron Age. This aspect of the West Norwegian graves marks a distinct disparity with the Icelandic graves in this material, which never lie close to farm cores or on the home-fields (the distance between farm cores and graves is addressed in the section on distance analysis below). In contrast to the West Norwegian graves, all the Icelandic graves in the material are situated near routes, and most of them are also near farm boundaries. This situation is rare in the West Norwegian Iron Age material, and almost non-existent in the Late Iron Age. There are two sets of attributes that appear to complement each other, which should be discussed together. The first set consists of graves on a beach, and/or overlooking water, and/or in a welcoming position, i.e., on the track between the boat-landing and the farm core. In West Norway, these locations are common both for Late Iron Age graves and for Iron Age graves in general. The attributes all have to do with water in some form or other. It is tempting to suggest that they indicate travel by boat or on ice. The second set consists of graves overlooking a route, and in post-box positions, i.e., on the track leading from a main route to the farm core. This situation is typical for Iceland, but these locations 2010 B. Gjerland and C. Keller 165 are rare in the West Norwegian material. These attributes also have to do with travel, only this time on land, i.e., on foot or on horse-back. Travel on fjords and on lakes is traditional in West Norway, while travel on foot or on horseback is traditional in Iceland. Still, quite a few burial sites have a marine orientation. It is possible that the two groups of attributes represent similar connotations, only adapted to the local conditions for travel. Pagan Burials: Distance Analysis It was pointed out that the Iron Age burials in West Norway were situated closer to the farm cores than the Icelandic burials in this analysis. The West Norwegian graves have an affinity to farm cores and home fields, while the Icelandic graves have an affinity to farm boundaries and routes some distance away from the farm cores. For a proper comparison, the distances between graves and farm cores need to be specified (Table 2, Fig. 3). This study shows that the West Norwegian graves range from 70 to 700 m from the farm cores, but the main body of the material lies within a medium range of 100 to 300 m. By comparison, the Icelandic burial sites range from 300 to an impressive 1750 m, with the main body of the material in the range of 500 to 1200 m from the farm cores. Based on general knowledge of Icelandic burials, these examples may tilt towards the extreme, but distances in the vicinity of 500 m seem to be quite common (see Maher 2009 for extensive studies of graves and locations in Iceland). The general observation that there is a difference in the location of the pagan burials between West Norway and Iceland seems to be largely confirmed by the distance analysis, although caution must be urged due to the limited sample size. Pagan Burials: Discussion The differences between the West Norwegian and the Icelandic burial locations may be addressed in several ways. To what extent did the landscape of the dead resemble the landscape of the living? The attribute analysis suggests that in West Norway, burials from both the Early and Late Iron Age often were located near the farm core, i.e., close to where people lived, and close to the social arena. It is equally significant that the graves were situated at or in the direct vicinity of the home field. The farming economy of West Norway was a mixed economy; cereal cultivation was combined Figure 3. Distance between burial sites and farm cores in samples from Iceland and West Norway. Table 2. Distance between burial sites and farm cores in samples from Iceland and West Norway. Area Distance (m) West Norway 70 75 100 140 140 230 280 340 350 400 400 500 750 800 Iceland 300 500 500 500 500 1100 1240 1400 1750 166 Journal of the North Atlantic Special Volume 2 Icelandic culture for centuries (Adolf Friðriksson in Eldjárn 2000:610). Graham-Campbell and Batey (1998:145) observed that in Scandinavian Scotland the dead were often “… removed some distance from the living for burial …”, i.e., similar to the pattern in Iceland. They also noted (1998:145–156) that there was “a marked tendency in Scandinavian Scotland to utilize pre-existing mounds, thus ensuring the identification of the burial place in the landscape.” Using other peoples’ cemeteries may be a way of demonstrating power, but it may also be pure convenience (see Barrett 2003, Geake 2003, Meaney 2003, Richards 2003 for overview of Norse burial customs in Britain. Ambrosiani 1998 for Ireland, Andersson 2005 and Welinder 2003 for Sweden). In addition, there were Norse graves facing the sea, or lakes, which apparently is quite common in the rest of the Norse Atlantic area (Graham-Campbell and Batey 1998:145). The Christianization of Norway and Iceland Scandinavians visiting the Continent or the British Isles prior to the Viking Age would have encountered Christians and Christianity in many forms and places. The most likely contact areas were Germany6, Anglo-Saxon England7, Ireland, and Frisia8 (Sawyer, B. 1987; Sawyer, P.1987; Sawyer et al. 1987). The thirteenth-century Icelandic chronicler Snorri Sturlusson (Sturlusson et al. 1987) described the Christianization of Norway. Playing down the German influence as illustrated by Rimbert (Robinson 1921) and Adam of Bremen (Bremensis et al. 2002), he called attention to the Christianization efforts of Norway’s three missionary kings: Hákon Haraldsson “the good”9 (A.D. 933–959), Óláfr Tryggvason (A.D. 994–999), and Óláfr Haraldsson “the saint” (A.D. 1015–1028). Norwegian historians have traditionally embraced the Anglo-Saxon Church as “the mother of the Norwegian Church”10, in line with Sturlusson’s description (Keyser 1856–58; Maurer 1855–56, 1895; and especially Taranger 1890; see Myking 2001 for historiography; also Bang 1887, 1912; Birkeli 1982; Brendalsmo 2001; Bull 1912; Helle 1995a, b; Jackson 1994; Koht 1921; Kolsrud 1958; Kragh 1995; Lönroth 1963; Løwe 1995; Sigurðsson 2003a:29; Sveaas Andersen 1995). The charismatic kingships that appeared during the Viking Period developed to a medieval statelike structure under King Hákon Hákonarson (A.D. 1217–1263). The Gregorian Church Reform11 was launched in Europe in the eleventh century to disengage the Church from the power of secular leaders. Prior to the reform, most churches in Northern Europe were privately built and owned, either by the community, the nobility, or the kings. This was also the situation in Scandinavia. with animal husbandry and supplemented with hunting and fishing. Most indigenous religions in farming communities somehow involve ancestors, fertility, and productivity (Dumézil 1977; Steinsland 2005:77–81, 144–164). It makes sense, therefore, that so many of the West Norwegian graves are located on or are addressing arable land. It is tempting to interpret this close-to-home location as related to ancestral worship (see Steinsland 2005:327–357). Still, it is interesting that many of the graves in West Norway and Iceland seem to relate to the topic of travel, either because the burial location is facing the sea, or because the dead was buried in a boat (Müller-Wille 1970, Solberg 2000:263, Westerdahl 1991). Adolf Friðriksson wrote (in Eldjárn 2000:591–592) that graves in Iceland were usually associated with a farm, but he continued: “In contrast to the Christian graveyards, pagan burials were not situated by the farmhouses or even within the boundaries of the home field, but on the nearest suitable ground outside the hayfields; on a low rise or bank, perhaps half a kilometer or so away from the farmhouse.” He has also stated that in some cases graves were located near the boundaries between properties, and in other cases beside a road or a track (loc. cit. and Friðriksson 2004). Torun Zachrisson speaks of rune-stones marking boundaries in eleventh-century Sweden (Zachrisson 1998:194–196), and refers to the Irish practice of burials with ogam inscriptions marking boundaries (in Charles-Edwards 1976:85). The attribute analysis in Figure 2 supports such relationships. In addition, the comparison with West Norway points to the fact that there is an almost negative correlation between Icelandic burials, and farm cores and home fields. The colonization or “landnám” situation in Iceland may have required a different burial practice simply for legal reasons. When the first settlers arrived in Iceland around A.D. 874, there were no ancestors buried on the land, and hence no graves to worship. The Italian archaeologist Elena Garcea has stated (in a completely different context) that “by infusing the land with the remains of your people, you claim it” (quoted in Gwin 2008:137). It is possible that the graves of the early settlers in Iceland might have served as territorial claims. It is interesting that the graves both in Iceland and West Norway have an affinity to the subject of travel, either routes on land or water. This commonality may indicate a connection between Scandinavian and Icelandic burial customs, but Scandinavia is not the only possible source of influence; at the time of the colonization of Iceland, around A.D. 874, there were several Norse kingdoms in the British Isles (see Forte et al. 2005 for details). These Anglo- Saxon-Norse hybrid societies clearly influenced 2010 B. Gjerland and C. Keller 167 copal seat was established at Garðar in the Eastern Settlement (Keller 1991; Seaver 1996:33, 64–65). Church Building The contrasts between Norwegian and Icelandic medieval churches were striking. The oldest Norwegian churches were corner-post constructions in wood with standing “staves” (vertical timbers) dug into the ground. Post-holes from more than 30 such churches have been excavated under the floors of standing churches. Norway is famous for its stave-churches, which probably numbered some 750 by the end of the Middle Ages; today only about 60 remain. Their construction resembles that of the corner-post churches, but the vertical timbers all rest upon a frame of wooden sills laid out on the ground. So in contrast to the stone churches and the turf churches, stave churches leave no post-holes in the ground. Hence these church sites are difficult to identify archaeologically. (Christie 1981, 1983; KLNM Vol. 17:95– 107 “Stavkirke” [compare to KLNM Vol. 17:84–95 “Stavbygning”]—available online at http://www. stavkirke.org and http://www.stavechurch.org, respectively). The ca. 150 Norwegian stone churches that are still standing were built after ca A.D. 1100, most of them probably by foreign masons, and few if any stone churches were built after the Black Death A.D. 1349 (Ekroll 1997, Ekroll et al. 2000, Lidén 1991; see also Lidén 2008 [http://kunsthistorie.com/ wiki/index.php/portal:steinkirke]). It is virtually unknown that North Norway also had a number of turf churches. None are standing today, and very little is known about these churches, their construction, and dating (Bratrein 1968). In Iceland, there are written records about ca. 320 parish churches in the Middle Ages. In addition, there were numerous small churches that did not employ a priest (Vésteinsson 2000:93, 295–296). Many of these retained burial rights up towards the High Middle Ages. There are no standing churches from the Middle Ages in Iceland, but both wood- and turf-churches are known from the written records, and several turf churches have been excavated (KLNM Vol. 17:101– 104, Vol. 7:121–122; Kristjánsdóttir 2004:45–50, 119–149; Vésteinsson 2000:93–143). The Icelandic turf churches were basically timber buildings (both corner-post and stave-constructions have been excavated) with outer protective walls of turf and stone. The gable ends were made of wood, or at least the western gable which held the entrance. Turf houses require high maintenance and are not long-lasting. Contrary to Norway and Greenland, there are no medieval stone churches in Iceland. With the establishment of a Norwegian Church Province in Niðarós (present-day Trondheim) in A.D 1152–53, a true North Atlantic archbishopric was set up12 (Imsen 2003). The foundation document Canones Nidarosiensis established that the Gregorian Reform be implemented in the new church province (Bagge 2003:55–61, 80). With this reform, the proprietary church system in Norway was brought to an end, and private churches were transferred to the Church. This transfer included the many churches that belonged to the King. The intention was to apply the Gregorian Reform throughout the Niðarós Province, but after King Sverrir’s (A.D. 1177–1202) fall-out with the Church during the Norwegian Civil Wars, the reform was impeded. It was never implemented to the full in Iceland and Greenland. There were Christians among the early immigrants to Iceland, and missionaries also came from abroad. Iceland was ruled by chieftains who met at the National Assembly “Alþingi” and ruled by a common law. There was no central authority; the political power rested with the chieftains. During pagan times, the goðar were cult leaders (Lucas and McGovern 2007; see also Lucas 2009). When the question of Conversion to Christianity was put before the National Assembly ca A.D. 1000 (Aðalsteinsson 1971, Kristjánsdóttir 2004:142–152, Vésteinsson 2000:17–37), the goðar changed their religion, but maintained their influence on religious affairs. They resented intervention from the central clerical authorities, and the tension probably increased when Iceland and Greenland became parts of the Niðarós Church Province A.D. 1152–53. Bishops appointed by the Icelanders themselves sat in office until 1237, but the right to appoint bishops rested with the chapter in Niðarós from A.D. 1238 until the reformation 1536. Throughout Iceland, churches had been erected on private farms. As elsewhere, farmers donated property to the churches, and gradually two types of churches emerged: the “staðir” churches had full ownership of the farm at which they were located, and were financially attractive. The rest were called bændakirkjur (peasant churches). The struggle between the Church and the goðar in the twelfth and thirteenth centuries to control the “staðir-churches” is known as “staðamál” (Sigurðsson 2003b:122–124, Stefánsson 1995, Vésteinsson 2000:90–91). When Iceland and Greenland became parts of the Norwegian kingdom from A.D. 1262–64, the time of chiefly rule was brought to an end. However, as mentioned above, the Gregorian reform was never properly implemented in Iceland. Hence the proprietary church system prevailed until the reformation. Greenland was allegedly pagan for the first few years, but converted to Christianity around the same time as the Icelanders. Their first bishop, Arnald, was consecrated in Lund A.D. 1124, and the Epis168 Journal of the North Atlantic Special Volume 2 Greenland’s churches appear to be a mixture of Icelandic and Norwegian (or at least European) traditions. Unlike Iceland, Greenland had several European-style stone churches, particularly the near-intact Hvalsey Church in Qaqortukulooq built shortly after A.D. 1300 (Krogh 1982), which has been suggested to be very similar to Norwegian churches (Roussell 1941). The smaller churches, including the so-called Þjóðhild’s church at Brattahlið/ Qassiarsuk (Krogh 1982:27–52, Krogh 1983) appear to be in the same tradition as the Icelandic turf churches. Churches: Attribute Analysis A traveler to the Norwegian countryside will notice the white churches perched majestically on top of hills overlooking the farms below or standing proudly on the beaches along the west coast supervising the traffic on the fjords. Going to Iceland, the same traveler will be surprised to find rural churches nested neatly among the farm houses, not at all in such ostentatious locations as the Norwegian churches. Why this difference in church location between Norway and Iceland? Does it mean anything? The visual impact is striking, but can the difference be documented, and how? To have a closer look at how the churches are actually located, it is natural to start with an attribute analysis similar to the one made for the burials (Table 3, Fig. 4). To facilitate comparisons, the same eight attributes that were used in the burial analysis are utilized once more. In West Norway, 15 medieval church-sites in the Sunnfjord region were selected and visited for analysis— the same region and some of the same farms that Table 3. Landscape attributes connected to medieval church sites in West Norway, Iceland, and the Eastern Setlement and the Western Settlement in Norse Greenland. Attributes Area Farm core Home field Route Post-box Boundary Water Beach Welcome Raw data West Norway 5 12 12 0 2 13 8 1 Iceland 10 10 5 4 0 2 0 0 Greenland 13 13 5 3 0 12 6 7 Normalized data West Norway 9.4 22.6 22.6 0.0 3.8 24.5 15.1 1.9 Iceland 32.3 32.3 16.1 12.9 0.0 6.5 0.0 0.0 Greenland 22.0 22.0 8.5 5.1 0.0 20.3 10.2 11.9 Figure 4. Radar diagram showing landscape attributes connected to medieval church sites in West Norway, Iceland, and the Eastern Settlement and the Western Settlement in Norse Greenland. Eight attributes are indicated around the diagram. They are not mutually exclusive, i.e., one church site may feature one attribute, another two or three attributes. 2010 B. Gjerland and C. Keller 169 The church locations in these samples are strikingly different. The West Norwegian churches in the sample have an almost bimodal distribution, or a slanted figure of eight, in the graph (Fig. 4). Up and to the right are the attributes farm core, home field, and route. The churches with these attributes belong to what is called “Phase 1”, which will receive special attention in the discussion below. Down and to the left is another group of church sites that belongs to what is called “Phase 2”, which also is discussed below. The Icelandic churches in the sample have an overwhelming affinity to farm core and home field, similar to the Norwegian “Phase 1” churches. There is also some affinity to route and post-box locations. The Greenlandic churches in the sample have a great affinity to water and welcome positions. This is hardly surprising, given the fact that most Norse farms in Greenland are located near the shore, much more so than in West Norway, where there are also many inland farms. The Greenland churches also have a strong affinity to farm cores and home field, just like the medieval churches in Iceland. Churches: Distance Analysis Can the various church location types be better understood if their distances from the farm cores were visited during the pagan burial analysis. There are no longer medieval churches at these sites; most have early modern churches where the medieval churches once stood, and some have no church at all. The medieval churches in this district are all mentioned in the cadastre “Bergen Kalvskinn”13, and most of the medieval churchyards are still in use (Riksantikvaren14), although the buildings are more recent. Abandoned medieval church-sites have also been identified (Buckholm 1998). In Iceland, 10 medieval church sites were selected and visited for analysis, most of them in the Skagafjörður region in cooperation with Guðný Zoëga. The exception was the Hofstaðir site in the Mývatn region (Hofstaðir reports [http://www. instarch.is/instarch/midlun/netverkefni/arena/gogn/ hofstadir/]). The churches and dwellings are all identified archaeologically. In Norse Greenland, 13 church sites in the Eastern and in the Western Settlement were selected for analysis, the total being at least 18. None were visited for this analysis, but most of them are well known to one of the authors, and high-resolution documentation is easily available (Guldager et al. 2002, Krogh 1982). Churches and dwellings are identified archaeologically. Table 4. Distance between church and farm core in samples from West Norway, Greenland, and Iceland. Area Distance (m) Iceland 10 15 15 17 25 26 26 30 30 30 Greenland 0 5 15 20 25 30 30 35 40 45 50 80 140 West Norway 30 60 70 70 70 100 140 150 190 230 230 230 250 380 530 Figure 5. Distance between church and farm core in samples from West Norway, Greenland, and Iceland. In the Norwegian material, the dark blue bars are “Phase 1” churches, the rest are “Phase 2” churches. 170 Journal of the North Atlantic Special Volume 2 to 530 m. This arrangement probably has some very specific reasons which will be addressed below. The Norse Greenlandic churches are located between 0 and 140 m from the dwellings, i.e., in positions that place them between the Icelandic and the West Norwegian churches in the sample. Part of the reason may be that the houses on Norse farms in Greenland appear to be more dispersed than in Iceland. Churches: Discussion Both in the attribute analysis and in the distance analysis, the Icelandic churches stand out as being integrated among the farm houses. This relationship is characteristic for Icelandic churches even today. In the Middle Ages, the churches were surrounded by a churchyard, normally circular in shape. For the Icelanders, the construction of a cemetery among the farm houses implied a discontinuity in their burial customs: the furnished pagan graves some distance away from the farm core were replaced by a churchyard next door to the living (Table 5, Fig. 6). are measured? Table 4 and Figure 5 present the data and graphical representation of the distance in meters between medieval churches and their contemporary parent farms in West Norway, Iceland, and Greenland. In West Norway, the distance from the church is measured from the chancel to the center of the historical farm cores in land-reform maps. In Iceland, the distance is measured from the center of the church site, which is normally very small, to the center of either the (medieval?) dwelling or the center of the farm-mound. These locations are all based on archaeological finds such as house-ruins and observation of human bones indicating the church-yard. In Greenland, the distance is measured from the wall of the church ruin to the nearest wall of the dwelling, on the basis of survey maps. All the identifications in Greenland are based on archaeological recognition of the ruins. The Icelandic churches are characterized by lying close to the dwellings, in the range of 10 to 30 m. The Norwegian churches are located both nearby and far from the farm cores, ranging from 30 Table 5. Attribute analysis with location of pagan graves compared to medieval churches in the samples from Iceland. Attributes Material Farm core Home field Route Post-box Boundary Water Beach Welcome Raw data Pagan graves 0 0 8 3 6 7 7 7 Churches 10 10 5 4 0 2 0 0 Normalized data Pagan graves 0.0 0.0 21.1 7.9 15.8 18.4 18.4 18.4 Churches 32.3 32.3 16.1 12.9 0.0 6.5 0.0 0.0 Figure 6. Attribute analysis with location of pagan graves compared to medieval churches in the samples from Iceland. 2010 B. Gjerland and C. Keller 171 It is a likely assumption that the location of Icelandic churches reflects the proprietary church system, i.e., that the churches were originally privately built and owned. The most astonishing part is not that the early churches were built like this, but the fact that the tradition has prevailed until modern times. The West Norwegian church locations stand out (Table 6, Fig. 7). During the field survey, it was discovered that some of the churches were located almost like the churches in Iceland, i.e., close to home field and farm core, by a road, in what looked like private locations. These churches are tentatively labeled “Phase 1” churches. In the attributes analysis in Figure 4, the “Phase 1” churches are indicated top right; in the distance analysis in Figure 5, they are indicated by dark blue bars. The Norwegian churches are surrounded by modern church-yards which have expanded with the population; hence, they are most likely larger than in medieval times. This growth may explain the slightly greater distance in the Norwegian material between the “Phase 1” churches and the houses. In addition, the distance is measured to the center of the historical farm core as indicated by the historic maps. In both Iceland and Greenland, the distances are based on the archaeological material. The most striking church locations are the ones where the churches are located far from their parent farms. During the field survey, it was observed that many churches were lying by themselves, often at a beach or near a river, overlooking water. Some were placed on gravel estuaries15 in places convenient for boat landings. These church locations seem to be rather public in nature. Such churches are labeled “Phase 2” churches. In the distance analysis, there is no obvious gap to indicate the difference between the “Phase 1” and Table 6. Attributes analysis with graves and churches in the samples from West Norway. The location of Late Iron Age graves compared to the location of medieval church sites. Attributes Material Farm core Home field Route Post-box Boundary Water Beach Welcome Raw data Late Iron Age graves 8 8 0 1 0 5 5 7 Rest of Iron Age graves 8 11 8 3 1 10 8 5 Churches 7 12 12 0 2 13 8 1 Normalized data Late Iron Age graves 23.5 23.5 0.0 2.9 0.0 14.7 14.7 20.6 Rest of Iron Age graves 14.8 20.4 14.8 5.6 1.9 18.5 14.8 9.3 Churches 12.7 21.8 21.8 0.0 3.6 23.6 14.5 1.8 Figure 7. Attributes analysis with graves and churches in the samples from West Norway. The location of Late Iron Age graves compared to the location of medieval church sites. For background, graves from the rest of the Iron Age are included. 172 Journal of the North Atlantic Special Volume 2 “Phase 2” churches. The distance of 80 m between church and farm core has been used to distinguish between “Phase 1” and “Phase 2” West Norwegian churches, i.e., the “Phase 2” churches lie more than 80 m away from the farm core. This definition is also based on visual impression; the topography and the visual scale of the landscapes are important factors in such analyses. River estuaries served in the past as convenient places to load and unload cargo between boat and overland transport. Thus, estuaries became nodes in the communication systems along the coast. Bronze Age and Iron Age finds indicate that farms near the estuaries often became power-centers which controlled exchange networks in larger regions (Farbregd 1986). Middle Age towns were usually placed at waterways, and river estuaries were popular locations: both medieval Oslo and medieval Niðarós (Trondheim) were established at estuaries. Maybe the Phase 2 churches were not only meant to facilitate public access, but were built at strategic locations to serve as nodes in larger communication networks. Through the Middle Ages, the Roman Catholic Church developed into a well-organized, hierarchic administrative system for which communication was essential. The “Phase 1” and “Phase 2” churches may reflect different stages in the organizational development of the Church of Norway. “Phase 1” churches close to the farm core can easily be interpreted as private churches, also implying chronological differences where these churches could be oldest. The “Phase 2” churches may indicate public churchbuilding, whether the builder was a congregation, the Church, or the King. It is tempting to suggest that these churches may have been built after the ideas of the Gregorian reform had started to take hold. The churches could then belong to the second generation of medieval churches in West Norway, built in the twelfth century (see Lidén 1995:140). At the moment, it is difficult to talk about possible structural differences between the two phases. An analysis of the churches mentioned in the “Bergen Kalvskinn” cadastre indicates that in the first half of the fourteenth century the majority of these churches were parish churches (Tryti 1987:434–435). All the churches except one were stave churches, with the exception being a “Phase 1” stone church at Lunde in Gaular. Only four of the “Phase 2” churches were not parish churches. It is interesting that the Norse Greenland church locations end up in between the Icelandic and the West Norwegian churches, both in the attribute analysis and in the distance analysis. Similarly, the Greenland church buildings point in two directions: there are turf churches with circular churchyards like in Iceland, and there are stone churches with rectangular churchyards like in Norway and the rest of Europe. The late stone-churches in such high-status farms as Brattahlíð/Qassiarsuk, Hvalsey/Qaqortoqulooq, and Herjólfsnes/Ikigaat in the Eastern Settlement and Ánavík and Sandnes/Kilaarsarfik in the Western Settlement are all “Phase 2” churches facing public (?) boat-landings. However, the Cathedral at Garðar/Igaliku and the churches at Ketilsfjörðr/ Tasermiutsiaat and Uunatoq are integrated “Phase 1” churches in accordance with the “Icelandic” pattern, despite their size. It is worth recalling that all bishops in Iceland were Icelanders prior to A.D. 1238, while all the bishops in Greenland16 throughout the Middle Ages were Norwegians (see Seaver 1996). By the looks of it, the European-style stone churches may reflect a Norwegian influence on church building in Greenland. This may have several reasons, but it is relevant to ask whether Greenland ever experienced a struggle among the chiefs to control the churches, like the staðamál conflict among the goðar in Iceland. It is tempting to suggest that the Greenland Church developed along trajectories that were different from those of the Icelandic Church. Visual Landscape Analysis and Source Material The landscape analyses in this paper had a focus on finding methods to compare three very different landscapes in West Norway, Iceland, and Greenland. It set out with a range of observations and documentations. The landscape variables utilized in the end were selected because they revealed features characteristic for the sites. The attributes may well reflect the typical archaeological perception of sites in each of the landscapes. When this perception, however, is formalized, like in the eight attributes applied here, the ground for comparisons between different landscapes is laid. The attribute analyses needed to be backed up by measurements of distances between farm cores and graves, and between farm cores and churches. This gives a better understanding than indistinct statements of the type “the church lies close to the farm”. Within the realm of one landscape, such descriptive statements have their uses, but in a comparison between different landscapes, more precise information is needed. The analyses exposed the challenges that arise when archaeological and historical information from different countries is used in the same study. The diversity in the source material made it possible to identify, for example, Viking and medieval farm cores in the field, but due to the diverse nature of the source material, different criteria had to be applied in each sample area. In the present case, farm clusters were identified on the basis of the historical farms in 2010 B. Gjerland and C. Keller 173 in distinctly public places such as river estuaries and open beaches, i.e., traditional landing places for boats and cargo and clearly also social arenas, which may go way back in time. Tentatively, the West Norwegian churches in the sample have been classified as “Phase 1” and “Phase 2” churches. The “Phase 1” churches lie in locations similar to the Icelandic churches (which are all “Phase 1”) in that they are integrated in a house cluster, while the more public “Phase 2” churches are unknown in Iceland. Interestingly, Greenland has both “Phase 1” and “Phase 2” churches; the “Phase 2” churches are generally larger, later, and built in stone with angular church-yards. The different church-building patterns may somehow reflect differences in the legislation and/or the church organization, but also in social conventions. There may also be chronological differences, but the material does not yet lend itself to detailed chronological studies. There seems, however, to be a shift from church-building in the private sphere (the “Phase 1” churches) to church-building in the public sphere (the “Phase 2” churches). As mentioned before, it is tempting to associate the shift with the Gregorian Church Reform and its endeavors to abolish the proprietary church system. Norway was a kingdom or at times several kingdoms, while Iceland and Greenland were chiefly societies up to AD 1262–64. The building of the public “Phase 2” churches may also be a result of royal influence. The question of continuity versus discontinuity in the Norse sacred landscape cannot be given a simple answer. In the sample area in West Norway, the pagan graves and the “Phase 1” churches seem to represent continuity in their close proximity to the farm cores. The discontinuity appears with the introduction of the “Phase 2” churches, which are detached from the farm cores and are erected at more public places. Hence, the cult practice is also getting detached from the private sphere and shifted towards public arenas. Quite the opposite situation occurs in the sample area in North Iceland—the pagan graves are not associated with the house clusters, but frequently address rather public areas such as tracks and communication lines some distance away. These are graves and not buildings of worship, so there is no necessary reason to assume that their more public locations indicate public participation in cult at the site. The graves were meant to be seen and recognized by travelers, although today they have no surface markings. The intimacy of the Icelandic churches nested within the house clusters gives completely opposite connotations. The private character of these “Phase 1” churches is accentuated when compared to the public character of the “Phase 2” churches in West Norway. Churches are cult-buildings designed for the land-reform maps in West Norway, on the farmmounds with identified medieval characteristics in Iceland, and on the medieval ruins of buildings that were abandoned sometime between the twelfth and the fifteenth centuries in Greenland. The analysis has triggered a need to look more closely at the two different types of church locations, and to analyze them in a context of visual landscape analysis. Information available in West Norway about farm cores, home fields, and other structures offer unique opportunities for this sort of inquiry. Comparative analyses are inspiring and fruitful and will also be applied. Continuity and Discontinuity Was there such a thing as a sacred Norse geography in pagan and early Christian times? The limited regional studies in this paper indicate that there certainly were some fairly stereotypical conventions which were followed in each region, with a set of variations. Caution must be observed due to the small sample size and hence the danger of overrepresentation of local patterns. With this warning in mind, the following observations may give food for thought. First of all, the pagan burials in West Norway have a strong affinity to home field and lie close to farm core, while the similar burials in Iceland are located far from the farm cores. The Icelandic pagan graves lie close to routes and farm boundaries. This pattern appears to be fairly consistent even for a larger body of material (see Maher 2009). Thus, it may be fair to state that the burial customs in Iceland follow conventions which differ somewhat from those in West Norway. The fact that the Icelanders were in a land-taking process may have called for specific funerary practices to serve other needs than the ones back home, where the properties and settlements go way back in time. Alternatively, the Icelandic funerary practice may reflect influence from other communities; possibly with the Norse communities in the British Isles. Second, the churches in Iceland are without exception integrated in the medieval house clusters at the farm. This pattern must reflect the medieval church organization in Iceland, which was completely dominated by the proprietary church system, i.e., the churches were privately owned. However, the proprietary church system alone did not necessitate such locations, so there must have been additional, social conventions at work. The church locations in the West Norway sample area stand out in clear contrast to the Icelandic pattern: the distances between farm cores and churches vary considerably. A few churches are integrated among the farm cluster as in Iceland, but the majority of the churches lie far from the farm cores, often 174 Journal of the North Atlantic Special Volume 2 Barrett, J.H. 2003. Christian and pagan practice during the conversion of Viking Age Orkney and Shetland. Pp. 207–226, In M. Carver (Ed.). The Cross Goes North. Processes of Conversion in Northern Europe AD 300–1300. The University of York, York Medieval Press, York, UK. Benediktsson, J. (Ed.) 1986. Íslendingabók (by Ári Þorgilsson). Pp. 1–28, In Íslendingabók Landnámabók, Íslenzk fornrit, Vol. 1. Reykjavík, Iceland. Berg, A. 1968. Norske gardstun. Instituttet for sammenlignende kulturforskning. Serie B, Skrifter LV. Universitetsforlaget, Oslo, Norway. Birkeli, F. 1982. Hva vet vi om kristningen av Norge? Utforskingen av norsk kristendoms- og kirkehistorie fra 900- til 1200-tallet. Universitetsforlaget-Norwegian University Press, Oslo, Norway. Bratrein, H. 1968. Kirketuft-registreringer i Nord-Norge 1966–1968. Reports in the archives of Riksantikvaren, Oslo, Norway. Brendalsmo, J. 2001. Kirkebygg og kirkebyggere. Byggherrer i Trøndelag ca. 1000–1600. Ph.D. Dissertation. Norwegian University of Science and Technology (NTNU), Trondheim, Norway. Buckholm, M.B. 1998. Nedlagte kirker og kirkesteder fra middelalderen i Hordaland og Sogn og Fjordane. Arkeologiske avhandlinger og rapporter fra Universitetet i Bergen, Norway. Bull, E. 1912: Folk og kirke i middelalderen. Kristiania, Norway. Charles-Edwards, T.M. 1976. Boundaries in Irish law. Pp. 83–87, In P.H. Sawyer (Ed.). Medieval Settlement, Continuity, and Change. Crane Russak, and Company, New York, NY, USA Christie, H. 1981. Stavkirkene – arkitektur. Pp. 139–252, In K. Berg (Ed.). Norges Kunsthistorie Vol 1. Gyldendal, Oslo, Norway. Christie, H. 1983. Den første generasjonen kirker i Norge. [With English summary]. Hikuin No. 9. Moesgård, Højbjerg, Denmark. Dokset, O. 2007. Gloppen kommune—reguleringsplan for utviding av Sandane lufthamn, Hjelmeset gbnr. 54/1 og 3, Andenes gbnr. 55/7, - rapport frå arkeologisk registrering. Rapport frå Kulturavdelinga, Sogn og Fjordane fylkeskommune, Norway. Dumézil, G. 1977. Gods of the Ancient Northmen. University of California, Los Angeles, CA, USA. Ekroll, Ø. 1997. Med kleber og kalk. Norsk steinbygging i mellomalderen 1050–1550. Samlagets bøker for høgare utdanning, Oslo, Norway. Ekroll, Ø., M. Stige, and J. Havran 2000. Middelalder i Stein. Vol. 1. In J. Havran (Ed.). Kirker i Norge, Vol. 1. ARFO, Oslo, Norway. Eldjárn, K. 2000. Kuml og haugfé úr heiðnum sið á Íslandi. In A. Friðriksson (Ed.). 2nd Edition. Fornleifastofnun Ísland, Mál og menning, Þjóðminjasafn Íslands, Reykjavík, Iceland. Farbregd, O. 1986. Elveosar—gamle sentra på vandring. Spor. Fortidsnytt fra Midt-Norge. No. 2. DKNVS Museet. Trondheim, Norway. Forte, A., R. Oram, and F. Pedersen. 2005. Viking Empires. Cambridge University Press, Cambridge, UK. Friðriksson, A. 2000. Viking Burial Practices in Iceland. Pp. 549–610, In A. Friðriksson (Ed.). Kristján Eldjárn: Kuml og haugfé. Úr heiðnum sið á Islandi. 2nd Edition. Mál og Menning, Reykjavík, Iceland. social religious activity, and the location of the “Phase 1” churches underlines the private character of the Christian cult in Iceland. The present study is an attempt to address the issues of pagan and Christian cult in the Norse cultural landscapes. The comparative approach applied here does give the opportunity to observe both similarities and differences in the establishment and use of a symbolic landscape. Acknowledgments The paper is based on archaeological material from sample areas in three countries: Iceland, Greenland, and Norway. This broad effort would not have been possible without the cooperative attitude of numerous colleagues. First of all we will express our thanks to Guðný Zoëga of the Glaumbær Folk Museum and The Skagafjörður Church Project, who has willingly shared information and guided us in the field. Without her, this work could not have been done. Thanks also to Bryndís Zoëga of the same institution. Adolf Friðriksson and Garðar Guðmundsson of the Institute of Archaeology in Iceland (FSÍ) have supplied us with endless information about pagan burials in Iceland over the years, and Árni Einarsson of the University of Iceland (HÍ) and the Mývatn Research Station is constantly finding archaeological sites everywhere; they all deserve our gratitude. Thanks also to Ragnheiður Traustadóttir of the Hólar University College for her cooperation and support over many years. Literature Cited Bremensis, A., F.J. Tschan, and T. Reuter. 2002. History of the Archbishops of Hamburg-Bremen. Translated with an introduction and notes by F.J. Tschan and a new introduction and selected bibliography by T. Reuter. Columbia University Press, New York, NY. Aðalsteinsson, J.H. 1971. Kristnitakan á Íslandi. Reykjavík, Iceland. Ambrosiani, B. 1998. Ireland and Scandavia in the Early Viking Age: An archaeological response. Pp. 404–430, In H.B. Clarke, M. Ní Mhaonaigh, and R.Ó. Floinn (Eds.). Ireland and Scandinavia in the Early Viking Age. Four Courts Press, Dublin, Ireland. Andersson, G. 2005. Gravspråk som religiös strategi. Valsta och Skälby i Attundaland under Vikingatid och tidig medeltid. Riksantikvarieämbetet, Stockholm, Sweden. Bagge, S. 2003. Den heroiske tid – kirkereform og kirkekamp, 1153–1214. Pp. 51–80, In Imsen, S. (Ed.). Ecclesia Nidrosiensis, 1153–1537. Søkelys på Nidaroskirkens og Nidarosprovinsens historie. Tapir akademisk forlag. Trondheim, Norway. Bang, A.C. 1887. Udsigt over den norske kirkes historie under katholicismen. Christiania, Norway. Bang, A.C. 1912. Den norske kirkes historie med portrætter, faksimiler og billeder. Gyldendalske Boghandel, Nordisk Forlag, Kristiania, Norway. 2010 B. Gjerland and C. Keller 175 Koht, H. 1921. Medførte kristendommens innførelse et makttap for det gamle norske aristokrati? Originally published in: Innhogg og utsyn 1921. Republished in: A. Holmsen (Ed.). 1966. Norsk Middelalder. Utvalgte avhandlinger for historiestudiet. Universitetsforlaget, Norway. Pp 69–78. Kolsrud, O. 1958. Noregs kyrkjesoga. Vol. I. Millomalderen, Oslo, Norway. Kragh, C. 1995. Kristendommen og social og politisk endring i vikingtiden. Pp. 18–40, In A. Ågotnes (Ed.). Kristendommen slår rot. Onsdagskvelder i Bryggens Museum. Bryggens Museum, Bergen, Norway. Kristjánsdóttir, S. 2004. The Awakening of Christianity in Iceland. Discovery of a Timber Church and Graveyard at Þórarinsstaðir in Seyðisfjörður. Gotarc, Gothenburg Archaeological Thesis, Series B No 31. Gothenburg, Sweden. Kristjánsson, J. 1988. Eddas and Sagas. Iceland’s Medieval Literature. Translated by Peter Foote. Hið íslenska bókmenntafélag, Reykjavík, Iceland. Krogh, K.J. 1975. Den islandske kolonisation. K’ak’ortok’—Julianehåb 1775–1975. Julianehåb Museumsforening, Qarqortoq, Greenland. Krogh, K.J. 1982: Erik den Rødes Grønland. Nationalmuseets Forlag, Copenhagen, Denmark. Krogh, K.J. 1983. Gård og Kirke. Samhørighed mellem gård og kirke belyst gennem arkæologiske undersøgelser på Færøerne og i Grønland. Hikuin 9, Højbjerg, Denmark. Lidén, H.-E. 1991. Fra antikvitet til kulturminne: trekk av kulturminnevernets historie i Norge. Universitetsforlaget- Norwegian Univeristy Press, Oslo, Norway. Lidén, H.-E. 1995. De tidlige kirkene. Hvem bygget dem, hvem brukte dem og hvordan? Pp. 129–141, In Lidén, H.-E. (Ed). Møtet mellom hedendom og kristendom i Norge. Universitetsforlaget-Norwegian Univeristy Press, Oslo, Norway. Also available online at http://www.nb.no/utlevering/contentview. jsf?urn=URN:NBN:no-nb_digibok_2008070804070. Accessed 12 November 2010. Lidén, H.-E. 2001. Middelalderens stenarkitektur i Norge. In Norges kunsthistorie Vol. 2. [PROVIDE PUBLISHER], Oslo, Norway. Lidén, H.-E. 2008. Kirkene i Hordaland gjennom tidene. Gyldendal, Bergen, Norway. Lucas, G. (Ed.) 2009: Hofstaðir. Excavations of a Viking Age Feasting Hall in North-Eastern Iceland. Institute of Archaeology, Reykjavík, Iceland. Monograph No. 1. Lucas, G., and T. McGovern. 2007. Bloody slaughter: Ritual decapitation and display at the Viking settlement of Hostaðir, Iceland. European Journal of Archaeology 10(1):7–30. Also available on-line at http:// eja.sagepub.com/cgi/content/refs/10/1/7. Lynch, K. 1992. The Image of the City. 21st Edition. MIT Press, Cambridge, MA, USA. Lönnroth, L. 1963. Studier i Olaf Tryggvasons saga. Pp, 54–94, In Samlaren. Svenska Litteratursällskapet, Uppsala, Sweden. Løwe, G.C.V. 1995. Religionsskiftet i samspill med de politiske maktforholdene i Norge. En historiografisk undersøkelse. KULTs skriftserie nr. 72. Norges forskningsråd/ Norwegian Research Council, Oslo, Norway. Friðriksson, A. 2004. The Topography of Iron Age Burials in Iceland. Pp. 15–16, In G. Guðmundsson (Ed.). Current Issues in Nordic Archaeology. Proceedings of the 21st Conference of Nordic Archaeologists 6–9 September 2001 Akureyri Iceland. Society of Icelandic Archaeologists, Reykjavík, Iceland Gansum, T., G.B. Jerpåsen, and C. Keller. 1997. Arkeologisk landskapsanalyse med visuelle metoder. AmS Varia 28. Archaeological Museum in Stavanger. Stavanger, Norway. Geake, H. 2003. The Control of Burial Practice in Anglo- Saxon England. Pp. 259–269, In M. Carver (Ed.): The Cross Goes North: Processes of Conversion in Northern Europe AD 300–1300. The University of York, York Medieval Press, York, UK. Graham-Campbell, J., and C.E. Batey. 1998. Vikings in Scotland: An Archaeological Survey. Edinburgh University Press, Edinburgh, Scotland, UK. Gräslund, A.-S., and M. Müller-Wille. 1993. Gravskicket i Skandinavien under Vikingatiden. Pp. 186–187, In E. Roesdahl (Ed.). Viking og Hvidekrist. Norden og Europa 800–1200. 22. Europarådsutstilling, Copenhagen, Denmark. Guldager, O., S. Stummann Hansen, and S. Gleie. 2002. Medieval Farmsteads in Greenland. The Brattahlid region 1999–2000. Danish Polar Center, Copenhagen, Denmark. Gwim, P. 2008. Lost Tribes of the Green Sahara. National Geographic Magazine September 2008:126–143. Helle, K. 1995a. Under Kirke og Kongemakt 1130–1350. Aschehougs Norges Historie volume 3. Oslo, Norway. Helle, K. 1995b. Kongemakt og kristendom. Pp. 41–54, In A. Ågotnes (Ed.). Kristendommen slår rot. Onsdagskvelder i Bryggens Museum. Bryggens Museum, Bergen, Norway. Hreinsson, V. (Ed.). 1997. The Complete Sagas of Icelanders. Vols. I–V. Leifur Eiríksson Publishing, Reykjavík, Iceland. Imsen, S. 2003. Nidarosprovinsen. Pp. 15–45, In S. Imsen (Ed.). Ecclesia Nidrosiensis 1153–1537. Søkelys på Nidaroskirkens og Nidarosprovinsens historie. Tapir akademisk forlag, Trondheim, Norway. Iversen, F. 2004. Eiendom, makt og statsdannelse: Kongsgårder og gods i Hordaland i yngre jernalder og middelalder. Universitetet i Bergen, Bergen, Norway. Jackson, T.N. 1994. The Role of Óláfr Tryggvason in the Conversion of Russia. Pp. 7–25, In M. Rindal (Ed.). Three Studies on Vikings and Christianization. KULTs skriftserie No. 28/RELIGIONSSKIFTET No. 1. The Research Council of Norway, Oslo, Norway. Keller, C. 1991. The Norse in the North Atlantic. A model of Norse Greenlandic Medieval society. Pp. 126–141, In G.F. Bigelow (Ed.). The Norse of the North Atlantic. Acta Archaeologica vol. 61. Copenhagen, Denmark. Keyser, R. 1856–58. Den norske Kirkes Historie under Katholisismen. Vols. I–II. Tønsbergs Forlag, Christiania, Norway. KLNM – Kulturhistorisk Leksikon for Nordisk Middelalder fra vikingtid til reformasjonstid. 1956–1968. Vols. 1–21. Facsimile 1980 by Rosenkilde og Bagger, Copenhagen, Denmark. 176 Journal of the North Atlantic Special Volume 2 Sawyer, B., P. Sawyer, and I. Wood (Ed.s) 1987: The Christianization of Scandinavia. Report of a Symposium held at Kungälv, Sweden 4–9 August 1985. Viktoria Bokförlag, Alingsås, Sweden. Sawyer, P. 1987. The process of Scandinavian Christianization in the tenth and eleventh centuries. Pp. 68–87, In B. Sawyer, P. Sawyer, and I. Wood (Eds.).The Christianization of Scandinavia. Report of a Symposium held at Kungälv, Sweden 4–9 August 1985. Viktoria Bokförlag. Alingsås, Sweden. Schumacher, J. 1987. Kirken i middelaldersamfunnet. Series Kristendommen i Europa 2. Tano A.S., Oslo, Norway. Seaver, K.A. 1996: The Frozen Echo. Greenland and the Exploration of North America, ca. A.D. 1000–1500. Stanford University Press, Stanford, CA, USA. Sigurðsson, J.V. 2003a. Kristninga i Norden ca 750–1200. Det norske samiaget, Oslo, Norway. Sigurðsson, J.V. 2003b. Island og Nidaros. Pp. 121–140, In S. Imsen (Ed.) Ecclesia Nidrosiensis 1153–1537. Søkelys på Nidaroskirkens og Nidarosprovinsens historie. Tapir akademisk forlag, Trondheim, Norway. Sigurðsson, J.V. 2004. Innledning. Pp. 5–12, In J.V. Sigurðsson, M. Myking, and M. Rindal (Eds.). Religionsskiftet i Norden. Brytninger mellom nordisk og europeisk kultur 800–1200 e.Kr. Occasional papers Skriftserie 6/2004. Senter for studier i vikingtid og nordisk middelalder, Oslo, Norway. Sigurðsson, J.V., B. Gjerland, and G. Losnegård. 2005. Ingólfr. Norsk-islandsk hopehav. Selja forlag, Førde, Norway. Sturlusson, S., F. Hodnebø, L.M. Hollander, and K.A. Lie. 1987. Snorri: The Sagas of the Viking Kings of Norway: Heimskringla / with the original saga illustrations by Halfdan Egedius… et al. Translated by L.M. Hollander. F. Hodnebø and K.A. Lie (Eds.). Stenersen, Oslo, Norway. Sognnes, K. 1988. Sentrumsdannelser i Trøndelag. En kvantitativ analyse av gravmaterialet fra yngre jernalder. Pp. 59–65, In Fortiden i Trondheims bygrunn: Folkebibliotekstomten. Meddelelser Nr. 12, Trondheim, Norway. Solberg, B. 2000. Jernalderen i Norge: ca. 500 f.Kr. – 1030 e.Kr. 2nd Edition. Cappelen, Oslo, Norway. Stefánsson, M. 1995. Islandsk egenkirkevesen. In H.E. Lidén (Ed.). Møtet mellom hedendom og kristendom i Norge. Universitetsforlaget-Norwegian University Press, Oslo, Norway. Steinsland, G. 2005. Norrøn religion. Myter, riter, samfunn. Pax Forlag A/S, Oslo, Norway. Sveaas Andersen, P. 1995. Samlingen av Norge og kristningen av landet. 800–1130. 2nd Edition. First published in 1977 In K. Mykland (Ed.). Handbok i Norges historie, Scandanavian University Books. Universitetsforlaget , Bergen, Oslo, Norway. Taranger, A. 1890: Den angelsaksiske kirkes indflydelse paa den norske. Den norske historiske forening, Kristiania, Norway. Tryti, A.E. 1987. Kirkeorganisasjonen i Bergen bispedømme i første halvdel av 1300-tallet. Upublisert hovedfagsoppgave i historie, Universitetet i Bergen. Bergen, Norway. Maher, R.A. 2009. Landscapes of life and death: Social dimensions of a perceived landscape in Viking Age Iceland. Unpublished Ph.D. Dissertation. The City University of New York, New York, NY, USA. Available online at http://www.nabohome.org/postgraduates/ theses/ram/. Accessed 10 November 2010. Maurer, K. 1855–56. Die Bekehrung des norwegischen Stammes zum Christenthume: in ihrem geschichtlichen Verlaufe quellenmäßig geschildert. I–II. Kaiser, Munich, Germany. Maurer, K. 1895. Nogle Bemærkninger til Norges Kirkehistorie. Historisk Tidsskrift, Kristiania, Norway. Meaney, A.L. 2003. Anglo-Saxon pagan and early Christian attitudes to the dead. Pp. 229–241, In M. Carver (Ed.). The Cross Goes North: Processes of Conversion in Northern Europe AD 300–1300. The University of York, York Medieval Press, York, UK. Morris, R. 1989. Churches in the Landscape. Phoenix Giant, London, UK. Müller-Wille, M. 1970. Bestattung im Boot. Studien zu einer europäische Grabsitte. Offa 25/26. Myking, M. 2001. Vart Norge kristna frå England? Ein gjennomgang av norsk forsking med utgangspunkt i Absalon Tarangers avhandling Den angelsaksiske kirkes indflytelse paa den norske (1890). Centre for Viking and Medieval Studies, Oslo, Norway. Occasional papers 1/2001 Ólason, V. 1998. Dialogues with the Viking Age. Narration and representation in the Sagas of the Icelanders. [Original title: Samræður við söguöld.]Translated by Andrew Wawn. Heimskringla, Reykjavík, Iceland. Olsen, M. 1928. Farms and Fanes of Ancient Norway. The place-names of a country discussed in their bearings on social and religious history. H. Aschehoug and Co. (W. Nygaard), Oslo, Norway. Olsen, T.B. 2010. Jordbruksbosetning på Hjelmeset gjennom 4000 år. Arkeologiske undersøkelser på Hjelmeset, Gloppen k., Sogn og Fjordane. Prosjekt Sandane lufthavn og Hjelmeset lok. 7. Arkeologiske rapporter fra Bergen Museum nr. 1/2010. Bergen, Norway. Øye, I. (Ed.). 2002. Vestlandsgården – fire arkeologiske undersøkelser, Havrå – Grinde – Lee – Ormelid. Arkeologiske avhandlinger og rapporter 8 fra Universitetet i Bergen, Bergen, Norway. Richards, J.D. 2003. Pagans and Christians at a frontier: Viking burial in the Danelaw. Pp. 383–385, In M. Carver (Ed.). The Cross Goes North: Processes of Conversion in Northern Europe AD 300–1300. The University of York, York Medieval Press, York, UK. Robinson, C.H. 1921. Anskar, The Apostle of the North, 801–865. Translated from the Vita Anskarii by Bishop Rimbert his fellow missionary and successor. The Society for the Propagation of the Gospel in Foreign Parts, London, UK. Roussell, A. 1941. Farms and Churches in the Mediaeval Norse Settlements of Greenland. Meddelelser om Grønland vol. 89 (I). Copenhagen, Denmark. Sawyer, B. 1987. Scandinavian conversion histories. Pp. 88–110. In B. Sawyer, P. Sawyer, and I. Wood (Eds.). The Christianization of Scandinavia. Report of a Symposium held at Kungälv, Sweden 4–9 August 1985. Viktoria Bokförlag, Alingsås, Sweden. 2010 B. Gjerland and C. Keller 177 11Named after Pope Gregorius VII (A.D. 1073–1085), known from his showdown with Henry IV in Canossa A.D. 1077. The Gregorian Reform with its slogan “libertas ecclesiae’ was developed in Cluny, west of the Rhine, in a feudal context (Schumacher 1987:67) 12From A.D. 1103, Lund (at the time under Denmark, but located in present-day Sweden) had taken over as the Archbishopric of the Nordic countries. The Archbishopric of Niðarós was constructed by transferring some of the area under the Archdioceses of Lund and York to Niðarós, thus joining the Norse North Atlantic under a single, administrative unit. It consisted of Norway, Iceland, Greenland, Faroes, Orkney, and Man and the Isles, areas where Old Norse was spoken or understood. 13“Bergen Kalvskinn”: Cadastre for the Bergen diocese from the mid-fourteenth century (http://gandalf.aksis. uib.no/menota/prosjekter/kalvskinnet.html). 14The sites have been evaluated by the Riksantikvaren (http://www.riksantikvaren.no/English/) and are in a special registry. 15Sand or gravel estuaries like these are called “øyr” or “ør” in Norwegian, “eyri” in Icelandic, and “eyrr” in Old Norse. They are a common compound in place-names. 16The Bishops of Garðar in Greenland started with the consecration of Arnaldr in A.D. 1124 and ended with the death of Alf in A.D. 1378 (see Seaver 1996:33, 111–112 and Index “Bishops”). Bishops of Garðar and Bishops of Greenland were consecrated on a regular basis in Niðarós long after all contact with Greenland hand been broken in the early 15th century, but these were honorary titles and the holders never went to Greenland. Vésteinsson, O. 2000. The Christianization of Iceland. Priests, Power, and Social Change 1000–1300. Oxford University Press, Oxford, UK. Welinder, S. 2003. Christianity, politics, and ethnicity in Early Medieval Jämtland, mid Sweden. Pp. 509–530, In M. Carver (Ed.): The Cross Goes North: Processes of Conversion in Northern Europe AD 300–1300. The University of York, York Medieval Press, York, UK. Westerdahl, C. 1991. Norrlandsleden. The maritime cultural landscape of the Norrland sailing route. Pp. 105–120, In O. Crumlin-Pedersen (Ed.). Aspects of Maritime Scandinavia AD 200–1200. Proceedings of the Nordic Seminar on Maritime Aspects of Archaeology, Roskilde, Denmark, 13–15 March 1989. Zachrisson, T. 1998. Gård, gräns, gravfält. Sammanhang kring ädelmetalldepåer och runstenar från vikingatid och tidigmedeltid i Uppland och Gästrikland. Stockholm Studies in Archaeology 15. Stockholms Universitet. Stockholm, Sweden Endnotes 1To avoid linguistic confusion, the technical term “farm core” is used. It indicates the house cluster of a farm, i.e. the dwelling and economy buildings. The Nordic term “tun” is ambiguous (West-Nordic “tún” = fenced in field, cfr. Anglo-Saxon “tún” = fence; German “zûn” = wicker fence; modern English “town”). It may mean the area between the houses on a farm, i.e., a farm-yard, but in regions where buildings were far apart, the meaning of “tún” has expanded so it includes the home field (as in modern Icelandic) or even cultivated fields in general (KLNM vol 19:39–51). In certain parts of Viking Period Sweden and in Anglo-Saxon England, compound placenames with “-tun”/ “-tuna” may be linked to royal estates (Morris 1989:143). 2Farms consisting of several holdings are termed differently in Scandinavian languages; Swedish “by”, Norwegian “mangbølt gård” and “klyngetun” which also characterize the physical lay-out: a cluster of houses. 3Trøndelag is the region around the city of Trondheim and the Trondheimsfjord in central Norway. In the Middle Ages Trondheim was known as Niðarós, the Center of the Norwegian Archbishopric. 4Allegedly, the first colonizer of Iceland, Ingólfr Arnarson came from Dalsfjord in Sunnfjord, and landed in Iceland ca. A.D. 874 according to Ari Þorgilsson fróði’s Íslendingabók (Book of Icelanders) which was written ca. A.D. 1130 (Benediktsson 1986:5). 5In West Norway, this track is often called “sjøavegen”— “the road to the sea”. 6Scandinavia originally belonged to the Archdiocese of Hamburg-Bremen. 7Norwegian liturgy and Christian Law show Anglo-Saxon influence. 8Frisia was essential in instigating the North Sea trade. 9King Hákon Haraldsson “the good” was raised by King Athelstan of England, and returned to Norway a Christian. His kingdom was mostly West Norway. 10The Anglo-Saxon connection being brought about by the association of the Norwegian missionary kings with England, even though King Ólafr Haraldsson eventually was unseated and killed by the associates of King Cnut the Great.