W AS E RIK THE R ED ’S B RATTAHLIÐ L OCATED AT Q INNGUA ? A D ISSENTING V IEW 1 Kevin J. Edwards, J. Edward Schofield, and Jette Arneborg Introduction An account of Brattahlið without the stories about Erik the Red, his wife Tjodhilde, their son Leif the Lucky, the introduction of Christianity to Greenland and the journeys to Vinland is unthinkable. Brattahlið was where these events either started or took place — if we are to believe the authors of the Icelandic sagas. (Arneborg 2006, 14) B ut what if the ‘unthinkable’ is applied to the location of the iconic Brattahlið itself? It may not matter to the saga-derived narratives, but has the considerable investment of archaeological effort, national pride, and tourist exploitation been misplaced? Is it founded upon a convenient misrepresentation? The Leverhulme Trust and the Danish Ministry of Science Technology and Innovation are thanked for funding which permitted the research reported here to be accomplished. The comments of Christian Keller on a draft of the paper were much appreciated. Maps were drawn by Alison Sandison. Kevin J. Edwards (kevin.edwards@abdn.ac.uk) is professor in Physical Geography and adjunct professor in Archaeology, School of Geosciences, University of Aberdeen; J. Edward Schofield (j.e.schofield@abdn.ac.uk) is lecturer in Geography, School of Geosciences, University of Aberdeen; Jette Arneborg ( Jette.Arneborg@natmus.dk) is curator and senior researcher, Danish Middle Ages and Renaissance, The National Museum of Denmark, Copenhagen. Abstract: The location of Eric the Red’s farmstead of Brattahlið in Greenland’s Eastern Settlement has long been debated. Following investigations in the eighteenth and nineteenth centuries, it was later concluded that it lay in the modern settlement of Qassiarsuk. A contrary view has been propounded by Ole Guldager who has suggested that a Norse ruin group at Qinngua, at the top of Eiríksfjörðr (Tunulliarfik fjord), is a more likely location. This paper presents new palaeoenvironmental evidence involving pollen analysis and landscape history, together with a consideration of settlement structure culminating in the excavation of a putative church site, and suggests that wherever Eric’s farm was located, it was probably not at Qinngua. Keywords: Erik the Red, Brattahlið, Greenland, Norse, settlement, archaeology, palaeoenvironment Viking and Medieval Scandinavia 6 (2010) 83–99. 10.1484/J.VMS.1.102137 84 Kevin J. Edwards, J. Edward Schofield, and Jette Arneborg After various attempts to locate Brattahlið in the eighteenth and nineteenth centuries, it was later concluded that Eric the Red’s farmstead lay in the modern settlement of Qassiarsuk (Figure 1; Nørlund 1929, 28; Nørlund and Stenberger 1934). A contrary view, however, has been propounded by Ole Guldager (2000, 2002) and Guldager and others (2002); namely, that a Norse ruin group at Qinngua, at the top of Eiríksfjörðr (Tunulliarfik fjord), is a more likely location. This paper presents new palaeoenvironmental and archaeological evidence from the area, and suggests that wherever Eric’s farm was located, it was probably not at Qinngua. Background Viking-Age settlers left their homelands in western Norway in the ninth century AD, doubtless for a variety of reasons (cf. Fitzhugh and Ward 2000; Barrett 2003, 2008). The Norse diaspora across the North Atlantic Ocean embraced parts of the British Isles and the Faroe Islands before reaching Iceland around AD 874 (Figure 1). The settlement (landnám) of Iceland was undertaken by independent chieftains and their followers (Orri Vésteinsson 1998, 2000; Byock 2001). If the medieval Icelandic literature is to be taken at face value, Eiríkr Þorvaldsson (Eiríkr Rauði, ‘Erik the Red’) was amongst the descendents of these landnámsmenn — an apparently charismatic, if quarrelsome, individual who was subsequently banished from Iceland for three years for his involvement in a murder or manslaughter (his father Þorvaldr had been forced to leave Norway for a similar crime). According to the Sturlubók Landnámu (cf. Ólafur Halldórsson 1978, 74), Erik had already reconnoitred Greenland for potential settlement locations prior to the Greenlandic colonization of c. AD 985. Having done a selling job on the supposedly verdant new land, Erik chose a location within what became Eiríksfjörðr (Tunulliarfik fjord; Figure 1) for his homestead, which he called Brattahlið (‘the steep slope’). As befits (presumably) the most powerful man in Greenland during the landnám period, his farm might reasonably be expected to have comprised one of the largest areas of cultivable land in the Eastern Settlement (ON Eystribyggð). The latter is in the extreme south of Greenland, a sea journey of some 600 km from the other major locus of occupation, the Western Settlement (ON Vestribyggð). The Eastern Settlement was a more extensive area than the Western Settlement, with a more attractive climate, and it was closer to the sea lanes of Iceland and Europe. The location of Brattahlið and other places was problematic from an early stage. The two key written sources to the topography of the Eastern Settlement — Ívar WAS ERIK THE RED’S BRATTAHLIÐ LOCATED AT QINNGUA? 85 Figure 1. Part of the Norse Eastern Settlement of Greenland with places mentioned in the text. The inset shows southern Greenland and its location in the North Atlantic Ocean. Contours in metres. Bárdsson’s description of Norse Greenland dating from the middle of the fourteenth century (Finnur Jónsson 1930; Ólafur Halldórsson 1978, 133–37), albeit recorded within seventeenth- century Danish transcripts, and a list of churches and bishops in the Flateyjarbók written in the period c. 1387–94 (Guðbrandur Vigfússon and Unger 1860–68) — agree on neither the number of churches in Eiríksfjörðr nor their names. Bárdsson mentions Diurenes (Dýrnes) church situated to the left when entering Eiríksfjörðr, with the churches of Sólarfjöllum and then Hlíðarkirkja (or Brattahliðar) associated with Brattahlið, further up-fjord. The Flateyjarbók mentions two churches in Eiríksfjörðr — Hardsteinabergi and Brattahlið. The churches are the main fixed points combining the literary narratives with the actual landscape — provided, of course, that the churches could be found. From the beginning of the nineteenth century much effort was put into locating and identifying churches, with the additional goal of discovering the precise location of Brattahlið. Poul Nørlund (1929, 26–31) and Ole Guldager (2002) record 86 Kevin J. Edwards, J. Edward Schofield, and Jette Arneborg how Heinrich Peter von Eggers (1793), following surveys of Norse ruin sites by Aaron Arctander (1777–79 [1944]) and Andreas Bruun (1777–79 [1944]), had placed Brattahlið in the modern settlement of Igaliku (now considered to be the former episcopal centre of Garðar), and this was accepted by J. J. A. Worsaae (1845). In a letter dated 1840, however, the Reverend Jørgen Frederik Jørgensen (1841), priest of Julianehåb (the modern Qaqortoq), had suggested that Brattahlið lay at Qassiarsuk, but Det kongelige nordiske Oldskriftsselskab (the Royal Society of Northern Antiquarians), to whom the letter was addressed, failed to take up this proposition. Erroneously, Nørlund (1929, 27) and Nørlund and Stenberger (1934, 16 n. 4) say that the Norwegian architect Herman Schirmer (1886) was the first to suggest the Qassiarsuk/Brattahlið link. Surveys of farm sites by Gustav Holm Figure 2. Norse ruin groups around Qassiarsuk; the location of the approximate shoreline for (1883) and Daniel Bruun (1895) led AD 1000 is after Mikkelsen and others (2008). Finnur Jónsson (1898) to propose that Contours in metres. Brattahlið was more likely to be found at Qassiarsuk which featured the remains of a large farm (designated as ruin group Ø29 and Ø29a — Bruun 1895; Nørlund and Stenberger 1934). Following excavations at Ikigaat/Herjolfsnes in 1921 (Nørlund 1924) and Garðar in 1926 (Nørlund 1929), Poul Nørlund and Mårten Stenberger (1934) undertook investigations at Qassiarsuk in 1932. This revealed a large farm with a church. After some initial doubts (Nørlund and Stenberger 1934, 14–15) two additional ruins (at Qorlortoq (Ø33) and Qorlortup Itinnera (Ø35)) were located and identified as churches in the Qorlortoq Valley just north of Qassiarsuk during the 1932 field season. Together with Dýrnes Church, this gave four churches in the fjord. It was concluded that Qassiarsuk was the location of the Norse Brattahlið. That hypothesis has been generally accepted up to the present — locally, nationally, and touristically (Cooper and others 2005; UNESCO 2010). WAS ERIK THE RED’S BRATTAHLIÐ LOCATED AT QINNGUA? 87 Figure 3. View south-east across part of the Norse farm complex at Qassiarsuk (ruin group Ø29a) in 2004: (A) church within its rectangular wall; (B) probable storehouse; (C) byre-barn complex. (Photograph by K. J. Edwards.) The claims for Qassiarsuk (Figures 2 and 3) are based on a mix of structures (e.g. dwellings, byres, storehouses, pens, churches, graveyards, possible thing booths, field walls, midden); its location in the sheltered inner fjord with good agricultural land; and its approximate adherence to saga testimony (this last is not necessarily reliable). What might be considered lacking is a noteworthy incline or hill that would justify the name Brattahlið (‘the steep slope’), if that really was a major topographical feature. It is possible, of course, that the name was brought from Erik’s Icelandic or Norwegian homeland. In two stimulating articles, Ole Guldager (2000, 2002) has questioned this orthodoxy and advances a claim for Qinngua (ruin group Ø39) at the head of Tunulliarfik fjord as a more likely location for Brattahlið. In summary, Guldager points out that the size and number of ruins in Qinngua are much greater than those at Qassiarsuk, and that Ø39 contains an unexcavated church ruin which is central to his hypothesis (Guldager and others 2002, 45). 88 Kevin J. Edwards, J. Edward Schofield, and Jette Arneborg Guldager’s Hypothesis Tacitly following Nørlund and Stenberger (1934) and Roussell (1941), Guldager (2002, 79) sees the important criteria for determining a major farm site as (1) a favourable location, (2) a large farm, and (3) the presence of a church. The first criterion is deemed to be difficult to evaluate because of landscape and environmental changes (e.g. in climate, sea-level, and vegetation) since Norse abandonment of the Eastern Settlement around the late fifteenth century AD. The criterion of farm size is contentious. There is no certainty that the original landnám farms were the largest in any given area (cf. Inga S. Kristjönudóttir 2005) — and such a consideration contributed to the mistaken belief that Brattahlið might have been found at Igaliku/Garðar (see above). Nor is there much assurance that known ruins are strictly contemporaneous, although it might be thought reasonable to suppose that the number of ruins on a site could act as a possible benchmark when considering the status of a farm: successful farms expanded, less successful ones contracted. As Ívar Bárdsson observed, the lawman of Greenland was the chieftain at Brattahlið (Finnur Jónsson 1930) and he was perhaps unlikely to occupy a diminutive farm. Acknowledging such concerns, farm size can be assessed in several ways. Rather than using the number of single ruins within a farm complex as an indication of farm size (or indeed, factoring in the area of the associated homefield at least), Guldager agrees with Keller (1989) that the cumulative floor area of all farm buildings represents a better measure. The lack of dating for the 450 or so ruin groups (about half of which are farms as opposed to small sites and shielings) in the Eastern Settlement is problematic of course. As noted, there is no certainty of contemporaneity, although Guldager considers that ruins dating to the earliest phase of settlement are ‘quite indistinct and easy to distinguish from later ones’ (2002, 80–81). The significance of this differentiation for selecting a site as Brattahlið is not explored further. ‘Indistinct’ suggests unclear and therefore subject to being overlooked; while Brattahlið would have been an early site, it would likely show signs of multiperiodicity. Thirty-seven farm sites (many of which did not have ‘parish’ churches) had a total floor area exceeding two thousand square metres. The largest of these turns out to be at Qinngua — ruin group Ø39 — which is about twice the size (4652 v. 2295 square metres) of the hitherto favoured site (Ø29/29a) at Qassiarsuk. For Guldager, ‘The farm seems to consist of three holdings’ each with dwellings and byres — indeed, the large central one ‘could be two holdings’ (2002, 85). What is unclear is why the site at Qinngua is deemed to be one site when it could in fact be three (or four) separate ones (Figure 4). Using the same WAS ERIK THE RED’S BRATTAHLIÐ LOCATED AT QINNGUA? 89 Figure 4. Norse farm structures at Qinngua (simplified after Guldager and others 2002). logic, why would the southern farms on the Qassiarsuk plain (Ø28, Ø28a) not also be part of the same complex as Ø29/29a, not to mention the fell farm (Ø29b) (Figure 2)? If this were so — and assuming that Erik would have appropriated an extensive farming area for himself — then this greater Qassiarsuk entity would be considerably larger than Qinngua. It should be noted that many of the Norse farm buildings in Qassiarsuk have become severely degraded or have disappeared as a result of later farming activities. The third criterion — the presence of a church — is met for sixteen farms in the Eastern Settlement, many with a square church dyke and which may have functioned as parish churches. The latter are differentiated from the smaller churches or family chapels with their circular dykes. Unlike the large church farms, other large farms appear to have no church associated with them at all. Qinngua is of further interest to Guldager as it contains a presumed church (Figures 4–6). Such a structure had been mentioned by Clemmensen (1911), while Nørlund and Stenberger (1934, 18) not only noted it, but Nørlund conducted a small excavation which failed to find any conclusive proof as to function. Vebæk, who located many of the small churches, visited the site in 1949 without noticing 90 Kevin J. Edwards, J. Edward Schofield, and Jette Arneborg Figure 5. Structure 21 at Qinngua in 2010; the person is standing at the north-eastern corner, and the pollen site is circled. (Photograph by K. J. Edwards.) anything special (unpublished Vebæk diary 1949), and Krogh (1967, 1982) did not identify a church here. Albrethsen (1971) found charcoal to a depth of one metre inside a supposed dyke, but no human remains. Berglund (1980) cautiously referred to the supposed church as ‘a building inside a dyke’, and even raised doubts about the dyke, stating that the southern part was not present. It is difficult to see why the site can be deemed to denote ‘the presence of a church there’ (Guldager 2002, 86 and arguments therein). In spite of inconclusive evidence, it is maintained that So far, no graves have been found in Qinngua, but there can be only little doubt that this site is the remains of a church. The additional evidence is simply too strong […]. The presence of a possible church ruin in Qinngua makes this farm not only the largest in the country, but also a more likely candidate for the location of Brattahlið. (Guldager 2002, 88) Notwithstanding the lack of excavation evidence and the query raised above concerning the number of individual farms at Qinngua, Guldager prefers to see collections of holdings (as in the three farm units at Qinngua) as representing single farm complexes. On that basis, Brattahlið would be the combined group Ø29/29a (Figure 2) with its church (the Leiðar church (cf. Brattahliðar) as recorded by Ívar Bárdsson in the 1340s (Finnur Jónsson 1930)). WAS ERIK THE RED’S BRATTAHLIÐ LOCATED AT QINNGUA? 91 Figure 6. Plan of structure 21 at Qinngua (after Berglund 1980) showing the outline of Ole Guldager’s (2002) proposed church site (the dashed lines) and the 2002 excavation trenches (this paper). Excavations of the Presumed Church at Qinngua, 2002 In 2002, Jette Arneborg undertook limited excavation in connection with an ongoing research project that focuses on small churches in Norse Greenland. Three trenches (Figure 6) were excavated through the proposed church structure at Qinngua in an attempt to determine whether this feature does indeed represent the remains of a church surrounded by a churchyard and a churchyard dyke. Trench 1, measuring 1.5 x 8 m, was positioned on the southern side of the structure. The trench was placed so as to cut through the putative church building (northern part of the trench), the graveyard (centre of the trench), and the church dyke (southern part of the trench). The trench revealed a c. 50 cm deep sequence. At the base of the profile was yellow (natural) gravel. Immediately above this was a thin ‘landnám layer’, the standard Dano-Greenlandic archaeological term for the original land surface — typically a turf layer — encountered by the first Norse settlers. Above this a ‘cultural’ deposit containing charcoal was observed (the exception being the southern part of the trench where, instead, a ‘destruction layer’ of relict building turfs and stones was recorded). In the northern part of the trench 92 Kevin J. Edwards, J. Edward Schofield, and Jette Arneborg there were also large stones. According to Guldager’s survey these constitute material from the church fabric. Yet critically, no burials were found in the centre of the trench and, as Berglund (1980) predicted, neither was there the faintest trace of a dyke. In a Greenlandic context, all three elements — building, graves, dyke — are present at sites designated as churches. Trench 2, measuring 1 x 3.5 m, cut through the south-east corner of the supposed dyke from the ‘inside’ of the structure. Here the culture layer was about 40 cm thick. The landnám horizon, as described above, was absent, but a black, greasy layer mixed with abundant charcoal overlaid the basal gravel and larger stones. Above the black layer was a stratum of relict building turfs and stones. Again, no traces of either burials or a dyke were found. Trench 3, measuring 1 x 2.5 m, was positioned ‘outside’ the structure and abutting a single row of upright stones which supposedly represent the eastern side of the churchyard dyke. The row of stones is one of the few convincing structural features encountered during the excavation, and these were definitely placed for a purpose as they are wedged upright with smaller packing stones. This mode of construction is dissimilar to churchyard dykes found elsewhere in Norse Greenland which typically consist of two rows of large stones with a core of smaller stones, cut turfs, and soil (e.g. Nørlund 1929, 35, fig. 12). On the inside of the row of stones, the stratigraphy was similar to that described for the two other trenches, although human activity was more apparent judging by several thin layers of relict turfs, pieces of subfossil vegetation, and charcoal. The lack of burials and the fact that the supposed building was not surrounded by a dyke leads us to conclude that Guldager’s feature no. 21 was probably not a church. Although the excavations at Qinngua would seem to refute a key plank of Guldager’s argument, this does not mean that an undiscovered church does not exist within the settlement area — but it seems unlikely. Landscape Change When attempting to locate the large farms in the Eastern Settlement, Guldager (2002, 80) declines to discuss landscape and environmental changes, noting that this type of approach would be ‘difficult’. This does not, however, absolve their dismissal in connection with the task in hand. Research from two areas — sea-level change and palynology — may be elicited to inform the debate. WAS ERIK THE RED’S BRATTAHLIÐ LOCATED AT QINNGUA? 93 Sea-level Change Relative sea-level (RSL) change associated with glacio-isostatic adjustment in Greenland is well documented (e.g. Bennike and others 2002; Mikkelsen and others 2008). The Norse graveyard at Ikigaat/Herjolfsnes, famed for its driftwood coffins and clothed skeletons (Nørlund 1924) is now partly submerged (Berglund 1988) and Norse ruins at Igaliku were already close to or below sea-level in 1876 (Steenstrup and Kornerup 1881). Off Qassiarsuk, side-scan sonar studies within Tunulliarfik fjord reveal a drowned land surface to a depth of about 3 m below mean sea-level to a distance of 110 m from the present shoreline (Figure 2; Hoffmann and others 1999; Kuijpers and others 1999). It is estimated that the local community in the immediate area of this ‘official’ Brattahlið lost c. 50 ha of land to rising RSL between landnám and the Norse abandonment of the Eastern Settlement (Mikkelsen and others 2008, 48). Any consideration of changes in RSL and the bathymetry at the head of the fjord are constrained by the lack of detailed hydrographic data. It is apparent from the lowest astronomical tide-water depths of 5–9 m depicted on the hydrographic chart (1:80,000; Grønlands Vestkyst) at around 61°13’N (3 km south of Qinngua), that the upper end of the Tunulliarfik fjord shallows rapidly; there is also the tidal range with which to contend (currently c. 1.5–2.5 m at Narsarsuaq; WWW Tide and Current Predictor 2010). In addition, the river draining into the head of the fjord seems to carry a considerable quantity of glacial rock flour, and presumably other minerogenic material. The interplay of a lower RSL and the added shallowing from riverine sedimentary inputs may have made the area immediately offshore of present-day Qinngua un-navigable or even dry land in Norse times — even a distance of say 1 km from farm to shoreline may have seemed excessive to the first settlers. Is it conceivable that a Viking of Erik’s status would not be located very close to the coast? As it stands, the hydrographic evidence is of doubtful use in assisting the argument that Qinngua was the site of Brattahlið, although further survey could change this view. Palynology The importance of Brattahlið would lead to the assumption that a major farm was created at the start of landnám and that its existence probably continued until the abandonment of the Norse settlement in Greenland c. AD 1450. It was clearly a functioning farm when Ívar Bárdsson (or his agents) conveyed his report in the 94 Kevin J. Edwards, J. Edward Schofield, and Jette Arneborg mid-fourteenth century, and the mention of the lawman having resided there would seem to confirm its contemporary or previous status. Pollen and related radiocarbon data from an organic sediment section within Qinngua (Figures 4 and 5) reveal that occupation of the site (Figure 7, local pollen assemblage zone (LPAZ) QIN-2) lasted from c. AD 1020–1380 (before the record is truncated). Within the constraints of the 14C method, this could well indicate local settlement for the whole of the Norse period. (The hiatus at the QIN-2b/3 boundary suggests an AD 1380 date for the end of settlement is likely to be a minimum.) The pollen spectra show that the period of Norse settlement began with a small decrease in the pollen of tree and shrub birch and a slight increase in grasses, indicating a reduction in the extent of woodland and an expansion in grassland communities which would include any hay-producing homefields. The reduction in Betula pollen at the outset of subzone QIN-2a is only temporary, and above 41 cm (c. AD 1100) pollen curves for both birch species recover and approach the frequencies recorded prior to the Norse arrival (as seen in zone QIN-1). A corresponding decrease in grass pollen (representing a possible contraction in the area under grassland, unless animals were grazing away the flowering heads of grasses) is also evident. The enhanced representation for sorrels, buttercups, and horsetails is characteristic of pastoral infields and disturbed areas in present-day Greenland (Schofield and others 2007). The consistent presence, rather than increases, in sedges and willow suggests that the land was relatively well managed and that it was not allowed to degenerate. There is, though, a reduction in loss-on-ignition and the presence of sand layers, which are indicative of soil erosion within the catchment area of the site. The human dimension within zone QIN-2 is reinforced by the expansion in the charcoal curve reflecting fire incidence (perhaps from cooking and heating as much as land management) and dung-related fungi of which Sporormiella-type (HdV-113) is most characteristic (cf. van Geel and others 2003; Raper and Bush 2009). The patterns summarized above are not dissimilar to those found elsewhere within the Eastern Settlement (e.g. Fredskild 1973, 1978, 1988; Edwards and others 2008; Schofield and others 2008; Buckland and others 2009) other than in one respect — the persistence of birch (averaging >20 per cent with tree birch contributing >12 per cent of total land pollen) during the occupation period. In comparison, birch percentages in Fredskild’s (1978) QDA section from Qassiarsuk never exceed 5 per cent in the Norse-age deposits, while at Tasiusaq (Ø2), c. 12 km south-west of Qinngua, the birch woodland was destroyed following landnám and never became re-established (Edwards and others 2008). Given the high birch WAS ERIK THE RED’S BRATTAHLIÐ LOCATED AT QINNGUA? 95 Figure 7. Pollen diagram from Qinngua showing selected palynological, microscopic charcoal, and sedimentary data. The dates shown are modelled estimates based upon an age-depth model (Schofield and others 2010). + indicates <1% total land pollen (TLP). 96 Kevin J. Edwards, J. Edward Schofield, and Jette Arneborg percentages and their consistency throughout the Norse period, it seems reasonable to hypothesize that woodland management (e.g. coppicing, pollarding, or stand protection) was being practised at Qinngua — a responsible activity in the face of shortages of wood and timber. Whether a high-status farm such as Brattahlið would have given over sufficient area of land for management in this way is a moot point, and it would likely have attracted comment in the writings which refer to Erik’s farm as was seen elsewhere. Thus, when sailing up Einarsfjörðr (modern Igalikup Kangerlua) towards Garðar in the fourteenth century, Ívar Bárdsson had noted that on the east side of the fjord, ‘there lies a big forest, which belongs to the cathedral, and in which forest the cathedral has all its cattle both big and little’ (Bruun 1918, 182); while at the head of Eiríksfjörðr in the eighteenth century, Aaron Arctander recorded that a ‘considerable’ birch wood existed on the north-eastern side (Ingstad 1966, 50). The pollen data cannot be seen as decisive on this point, and it might be noted that the diagrams from Galium Kær, Comarum Sø, and Comarum Mose (Fredskild 1973), which are located inland and upslope of Qassiarsuk, have birch pollen percentages which approach those of Qinngua. The sites are, though, located away from Norse settlements, and Galium Kær and Comarum Sø are lake basins which have the potential to recruit pollen from wide areas; they cannot be seen as reflecting land use in intimate contact with farmlands in the way that the Qinngua and Qassiarsuk pollen profiles would do. Galium Kær, however, might indicate a local managed woodland area of a Brattahlið located at Qassiarsuk. Conclusions The investigation reported here casts doubt on Erik the Red’s Brattahlið being located at Qinngua. The farm construct, as posited by Guldager, is questionable, and crucially, subsequent excavation has revealed that the only proposed ecclesiastical structure is probably not a church. Furthermore, the topographic location of the site is perhaps suboptimal, and the record of vegetational history is unsupportive. Ole Guldager has advanced a hypothesis which has been tested by excavation, and in one of its key essentials that hypothesis has been rejected. Guldager’s paper, however, is welcome as it raises issues which should continue to be addressed. While this paper does not further a Brattahlið/Qassiarsuk link, it does not disprove such a connection. WAS ERIK THE RED’S BRATTAHLIÐ LOCATED AT QINNGUA? 97 Bibliography Primary Sources Albrethsen, Svend E. 1971. Beretning om Undersøgelse d. 18–19/6 1971, unpublished report, Copenhagen: National Museum of Denmark, Danish Middle Ages and Renaissance Berglund, Joel. 1980. Antikvariske Undersøgelser I Fåreholderdistriktet Sydgrønland, unpublished report, Copenhagen: National Museum of Denmark; Nuuk: Greenland National Museum & Archives Finnur Jónsson. 1930. Det Gamle Grønlands Beskrivelse af Ívar Bárðarson, Copenhagen: Levin and Munksgaards Grønlands Vestkyst, Julianehåb-Narssarssuaq, 1:80,000 (60°30’), 3rd edn, June 2000, Copenhagen: Kort & Matrikelstyrelsen Guðbrandur Vigfússon and Carl R. Unger, eds. 1860–68. Flateyjarbok: En samling af norske kongesaegar med indskudte mindre fortællinger om begivenheder i og udenfor Norge samt annaler, 3 vols, Oslo: Mallings Jørgensen, Jørgen F. 1841. Ruin Descriptions in South Greenland, unpublished letters, Copenhagen: National Museum of Denmark, Danish Middle Ages and Renaissance Secondary Sources Arctander, Aaron. 1777–79 [1944]. ‘Udskrivt af en Dagbog Holden i Grønland af Aaron Arctander, paa en Recognoscerings-reise i Julianæhaabs District i Aarene 1777–1779: Udg. H. Ostermann. Dagbøker av norddmenn på Grønland før 1814’, Norges Svalbard- og Ishavsundersøkelser, Meddelelser 58, 59–138 Arneborg, Jette. 2006. Saga Trails – Brattahlið, Garðar, Hvalsey Fjord’s Church and Herjolfsnes: Four Chieftain’s Farmsteads in the Norse Settlement of Greenland. A Visitor’s Guidebook, Narsaq: Vintervår Barrett, James H., ed. 2003. Contact, Continuity and Collapse: The Norse Colonization of the North Atlantic, Studies in the Early Middle Ages 5, Turnhout: Brepols Barrett, James H. 2008. ‘What Caused the Viking Age?’, Antiquity 82, 671–85 Bennike, Ole, Svante Björck, and Kurt Lambeckt. 2002. ‘Estimates of South Greenland LateGlacial Ice Limits from a New Relative Sea Level Curve’, Earth and Planetary Science Letters 197, 171–86 Berglund, Joel. 1988. Herjolfsnæs – Kirken ved Havet, Nanortalik Kommune, Qaqortoq: Qaqortoq Kommunes Forlag Bruun, Andreas. 1777–79 [1944]. ‘Dag-bog Holden paa Undersigning Reisen af Egnen ved Colonien Julianeshaab i Grønland, for 1777 og 1778: Udg. H. Ostermann. Dagbøker av Norddmenn på Grønland før 1814’, Norges Svalbard- og Ishavsundersøkelser, Meddelelser 58, 1–57 Bruun, Daniel. 1895. ‘Arkæologiske Undersøgelser i Julianehaabs Distrikt’, Meddelelser om Grønland 16, 171–461 Bruun, Daniel. 1918. ‘The Icelandic Colonization of Greenland’, Meddelelser om Grønland 57, 1–228 98 Kevin J. Edwards, J. Edward Schofield, and Jette Arneborg Buckland, Paul C., Kevin J. Edwards, Eva Panagiotakopulu, and J. Edward Schofield. 2009. ‘Palaeoecological Evidence for Manuring and Irrigation at Garðar (Igaliku), Norse Eastern Settlement, Greenland’, The Holocene 19, 105–16 Byock, Jesse L. 2001. Viking Age Iceland, London: Penguin Clemmensen, Mogens. 1911. ‘Undersøgelsesrejse i Sommeren 1910, Kirkeruiner fra Nordbotiden m.m. i Julianehaab Distrikt’, Meddelelser om Grønland 47, 285–358 Cooper, David, Rögnvaldur Guðmundsson, and Tom Muir. 2005. Destination Viking SAGALANDS: The Icelandic Sagas and Oral Tradition in the Nordic World, Reykjavík: Destination Viking SAGALANDS Edwards, Kevin J., J. Edward Schofield, and Dmitri Mauquoy. 2008. ‘High Resolution Paleoenvironmental and Chronological Investigations of Norse landnám at Tasiusaq, Eastern Settlement, Greenland’, Quaternary Research 69, 1–15 von Eggers, Heinrich P. 1793. ‘Om Grønlands Østerbygds Sande Beliggenhed’, Det Kongelige Landhusholdningsselskabs Skrifter 4B, 239–320 Finnur Jónsson. 1898. ‘Grønlands Gamle Topopgrafi efter Kilderne’, Meddelelser om Grønland 20, 265–329 Fitzhugh, William W., and Elisabeth I. Ward, eds. 2000. Vikings: The North Atlantic Saga, Washington, DC : Smithsonian Institution Press Fredskild, Bent. 1973. ‘Studies in the Vegetational History of Greenland’, Meddelelser om Grønland 198, 1–245 Fredskild, Bent. 1978. ‘Palaeobotanical Investigations of Some Peat Deposits of Norse Age at Qagssiarssuk, South Greenland’, Meddelelser om Grønland 204, 1–41 Fredskild, Bent, 1988. ‘Agriculture in a Marginal Area: South Greenland from the Norse Landnám (985 AD ) to the Present (1985 AD )’, in The Cultural Landscape: Past, Present and Future, ed. Hilary H. Birks, H. John B. Birks, Peter E. Kaland, and Dagfin Moe, Cambridge: Cambridge University Press, 381–93 van Geel, Bas, Janneke Buurman, Otto Brinkkemper, Jaap Schelvis, André Aptroot, Guido van Reenen, and Tom Hakbijl. 2003. ‘Environmental Reconstruction of a Roman Period Settlement Site in Uitgeest (The Netherlands), with Special Reference to Coprophilous Fungi’, Journal of Archaeological Science 30, 873–83 Guldager, Ole. 2000. ‘Hvor lå Brattahlid?’, Tidsskriftet Grønland 5, 161–72 Guldager, Ole. 2002. ‘Brattahlíð Reconsidered: Some Thoughts on the Social Structure of Medieval Norse Greenland, and the Location of Brattahlíð’, Archaeologia Islandica 2, 74–97 Guldager, Ole, Steffen Stummann Hansen, and Simon Gleie. 2002. Medieval Farmsteads in Greenland: The Brattahlid Region 1999–2000, Danish Polar Center Publications 9, Copenhagen: Danish Polar Center Hoffmann, Gerd, Antoon Kuijpers, Jörn Thiede, and Shipboard Party. 1999. ‘Climate Change and the Viking-Age Fjord Environment of the Eastern Settlement, South West Greenland: Cruise Report Poseidon Cruise No. 243’, Alfred-Wegener-Institute for Polar and Marine Research, Reports on Polar Research 333, 1–34 Holm, Gustav F. 1883. ‘Ruiner i Julianehaabs Distrikt’, Meddelelser om Grønland 6, 57–145 Inga S. Kristjönudóttir. 2005. ‘Hof and Hólar in Hjaltadalur: A Study of Central Farms in Hjaltadalur in Skagafjörður, Iceland in Viking and Early Medieval Times’, unpublished master’s thesis, Uppsala University, <http://www.holar.is/holarannsoknin/verkefni/ Inga_soley_Kristjonudottir_MA.pdf> WAS ERIK THE RED’S BRATTAHLIÐ LOCATED AT QINNGUA? 99 Ingstad, Helge. 1966. Land under the Pole Star, London: Jonathan Cape Keller, Christian. 1989. ‘The Eastern Settlement Reconsidered: Some Analyses of Norse Medieval Greenland’, unpublished doctoral thesis, University of Oslo, <http://www.nabohome.org/ postgraduates/theses/ck/> Krogh, Knud J. 1967. Viking Greenland, Copenhagen: National Museum of Denmark Krogh, Knud J. 1982. Erik den Rødes Grønland, Copenhagen: National Museum of Denmark Kuijpers, Antoon, Niels Abrahamsen, Gerd Hoffmann, Veit Hühnerbach, Peter Konradi, Helmar Kunzendorf, Naja Mikkelsen, Jörn Thiede, and Wilhelm Weinrebe. 1999. ‘Climate Change and the Viking Age Fjord Environment of the Eastern Settlement, South Greenland’, Geology of Greenland Survey Bulletin 183, 61–67 Mikkelsen, Naja, Antoon Kuijpers, and Jette Arneborg. 2008. ‘The Norse in Greenland and Late Holocene Sea-level Change’, Polar Record 44, 45–50 Nørlund, Poul. 1924. ‘Buried Norsemen at Herjolfsnes’, Meddelelser om Grønland 67, 1–207 Nørlund, Poul. 1929. ‘Norse Ruins at Gardar: The Episcopal Seat of Mediaeval Greenland’, Meddelelser om Grønland 76, 1–171 Nørlund, Poul, and Mårten Stenberger. 1934. ‘Brattahlid’, Meddelelser om Grønland 88, 1–161 Ólafur Halldórsson. 1978. Grænland í Miðaldaritum, Reykjavík: Sogufelag Orri Vésteinsson. 1998. ‘Patterns of Settlement in Iceland: A Study in Prehistory’, Saga-Book 25, 1–29 Orri Vésteinsson. 2000. The Christianization of Iceland: Priests, Power and Social Change 1000–1300, Oxford: Oxford University Press Raper, Diane, and Mark Bush. 2009. ‘A Test of Sporormiella Representation as a Predictor of Megaherbivore Presence and Abundance’, Quaternary Research 71, 490–96 Roussell, Aage. 1941. ‘Farms and Churches in the Mediaeval Norse Settlements of Greenland’, Meddelelser om Grønland 89, 1–354 Schirmer, Herman. 1886. ‘Om Beliggenheden af Gardar’, Norsk Historisk Tidsskrift 2.5, 412–17 Schofield, J. Edward, Kevin J. Edwards, and Charlie Christensen. 2008. ‘Environmental Impacts Around the Time of Norse landnám in the Qorlortoq Valley, Eastern Settlement, Greenland’, Journal of Archaeological Science 35, 1643–57 Schofield, J. Edward, Kevin J. Edwards, and J. Andy McMullen. 2007. ‘Modern Pollen-Vegetation Relationships in Subarctic Southern Greenland and the Interpretation of Fossil Pollen Data from the Norse Landnám’, Journal of Biogeography 34, 473–88 Schofield, J. Edward, Kevin J. Edwards, Tim M. Mighall, Antonio Martínez Cortizas, José Rodríguez-Racedo, and Gordon Cook. 2010. ‘An Integrated Geochemical and Palynological Study of Human Impacts and Climate Change from Southern Greenland’, Palaeogeography, Palaeoclimatology, Palaeoecology 295, 19–30 Steenstrup, Knud J. V., and Andreas Kornerup. 1881. ‘Expeditionen til Julianehaabs Distrikt i 1876’, Meddelelser om Grønland 2, 1–41 UNESCO. 2010. ‘Church Ruin at Hvalsø, Episcopal Residence at Gardar, and Brattahlid: A Norse/Eskimo Cultural Landscape’, <http://whc.unesco.org/en/tentativelists/1781/> (accessed July 2010) Worsaae, Jens J. A. 1845. Grønlands Historiske Mindesmærker, III, Copenhagen: Det kongelige nordiske Oldskrift Selskab WWW Tide and Current Predictor. 2010. ‘Tidal Height and Current Site Selection’, <http://tbone.biol.sc.edu/tide/> (accessed July 2010)