Eagle Hill Masthead

Journal of the North Altantic

    Aim and Scope
    Board of Editors
    Editorial Workflow
    Publication Charges

Co-published Journals
    Northeastern Naturalist
    Southeastern Naturalist
    Caribbean Naturalist
    Neotropical Naturalist
    Urban Naturalist
    Eastern Paleontologist
    Eastern Biologist

Eagle Hill Institute

About Journal of the North Atlantic

The Colonization of Iceland in Light of Isotope Analyses
Orri Vésteinsson and Hildur Gestsdóttir

Journal of the North Atlantic, Special Volume 7: 137–145

Full-text pdf (Accessible only to subscribers.To subscribe click here.)


Site by Bennett Web & Design Co.
Journal of the North Atlantic O. Vésteinsson and H. Gestsdóttir 2016 Special Volume 7 137 Introduction The firm dating of the landnám-tephra to the 870s, with reference to annual snow-melt in the Greenland icecap (Grönvold et al. 1995, Zielinski et al. 1997), liberated Icelandic Viking age archaeology from its fixation on the date of the start of colonization. For Kristján Eldjárn (1956), who laid the foundations for modern research into Viking age material culture inIceland, determining the timing of the initial colonization was a primary concern. It was also for his critics, most notably Margrét Hermanns-Auðardóttir (1989, 1991), who claimed a 7th-century date for the settlement of Iceland. As the contested archaeological deposits all post-date the tephra, the dispute evaporated with its definite dating. The resolution of this matter therefore effectively closed a debate that had occupied center-stage in Icelandic archaeology until then. This result coincided with a resurgence in Viking age archaeology in Iceland that has, in the past twenty years, produced great quantities of new data and, perhaps more importantly, several new lines of inquiry which are gradually changing our perceptions of this period in fundamental ways. One of these is the study of strontium isotopes in human bones, which allows immigrants to be distinguished from people born and bred in Iceland (Price and Gestsdóttir 2006, in review [this volume]). This research has profound implications both for our understanding of the rate and nature of immigration, and hence the colonization as a process, and for the value of burial data for demographic reconstruction. These themes are interlinked, and each needs to be considered to suggest plausible interpretations. Here we will start by discussing the evidence for immigration to Iceland in the 9th and 10th centuries before turning to the burial record. On this basis, it is possible to begin to piece together a more nuanced picture of the colonization as a social process where population movement is considered as one component in a story involving identity formation, community building, and political developments in a new society. Dating the Migration to Iceland In traditional scholarship, the chronological parameters for the colonization of Iceland were not in doubt: the first settlers arrived in A.D. 870 or 874, and the process was completed by A.D. 930 when a new chapter began with the establishment of the Alþing as the cornerstone of a new constitutional order. This formulation derives from Ari fróði, whose short chronicle on the history of Iceland, Íslendingabók, written in the A.D. 1120s, has long provided the framework for our understanding of the first 250 years of settlement in Iceland. All other written sources relating to this period are later, and none of them contain traditions that contradict or diverge from Ari’s account. The 13th-century and later Íslendingasögur (Sagas of Icelanders) and Landnámabók (the Book of Settlements) add a great quantity and variety of often colorful detail, but their concept of the colonization process is the same as Ari’s: there was a sixty-year period in which new arrivals could claim lands. There is a sense that those who came later within that period sometimes had to make do with less desirable land or oust earlier settlers from lands already occupied, but apart from that there is no conception in the medieval records of the settlement as a structured process; it was gradual for sixty years and then it ended. Those who came later could only acquire land through inheritance or purchase (Friðriksson and Vésteinsson 2003). This conception was first challenged by Ólafur Lárusson in an important paper published in 1929. Based on his examination of historical references to The Colonization of Iceland in Light of Isotope Analyses Orri Vésteinsson1,* and Hildur Gestsdóttir2 Abstract - A review of the mounting archaeological evidence for the colonization of Iceland suggests that the whole country was occupied within a couple of decades towards the end of the 9th century AD. Analyses of strontium in human bones show, however, that immigrants continued to arrive in Iceland throughout the 10th century. Here we discuss this apparent contradiction, suggesting that while continued immigration may have been needed to sustain the population, these patterns arise also from biases within the burial data. We argue that formal burial, of the kind that allows isotopic analyses, reflects growing affluence and the emergence of an indigenous gentry that sought to legitimate its power through association with the perceived homeland and its upper class. Viking Settlers of the North Atlantic: An Isotopic Approach Journal of the North Atlantic 1Department of Archaeology, University of Iceland, Reykjavik, Iceland. 2Institute of Archaeology, Reykjavík, Iceland. *Corresponding author - orri@hi.is 2016 Special Volume 7:137–145 Journal of the North Atlantic O. Vésteinsson and H. Gestsdóttir 2016 Special Volume 7 138 settlements in a valley in West-Iceland, he argued that whatever constitutional watershed may have been reached in 930, the actual occupation of the land took much longer and farms were being established throughout the 10th century, with the last foundations occurring as late as the 12th century (Lárusson 1929:334–342). Although he did not elaborate on this, it seems that Ólafur thought that there had not been enough immigrants coming between A.D. 870 and 930 to establish and maintain the ~4000 farms attested in later records. The post-930 foundations were therefore the result mainly of a natural increase in the already existing population, although he also reckoned with some immigration. Ólafur Lárusson’s thesis, although intellectually sound, was never influential; however, it gained importance as a foil to archaeological evidence beginning to emerge in the late 1990s. In contrast to Ólafur Lárusson’s model, the archaeological evidence suggested a remarkably rapid and complete occupation of the Icelandic landscape in the late 9th century (Vésteinsson et al. 2002:105–106). In recent years, more evidence has been coming to light supporting this picture. The most solid piece of evidence comes from the site of Sveigakot in the region of Mývatnssveit in northeastern Iceland, where extensive archaeological investigations have been underway since the early 1990s (overview in McGovern et al. 2007). Sveigakot is a low-status site located on marginal land that can never have been an ideal place to settle. Not only was the land poor—some wet meadow but mainly bare lava field—but also quite small, judging from distances to nearest neighbors with Viking age dates (see reconstruction in Thomson and Simpson 2007). The limited access to resources is reflected in the small size of the dwellings at the site. Sveigakot is the sort of place which common-sense, as well as Ólafur Lárusson’s model, would have dictated was established at the very end of the settlement process. It probably was, only this process seems to have taken place extremely fast. In Sveigakot, a tephra layer dated to ~A.D. 930–940 AD with reference to sedimentation rates in Lake Mývatn (Sigurgeirsson et al. 2013) overlies the earliest phase of settlement. This phase consists of a byre, which could have sheltered as many as 16 head of cattle, and a small sunken dwelling, measuring 4 m x 5 m with an adjacent outdoor activity area with two fireplaces. Both structures were abandoned and completely collapsed when the ~A.D. 930–940 tephra fell, suggesting that the abandonment occurred decades rather than years prior to the eruption. The dwelling had seven different floor layers and saw significant structural modifications in its long lifetime (Fig. 1). The midden associated with this earliest phase began to form directly on top of the landnám tephra of A.D. 871 ± 2. Taking all this evidence into account, it has been estimated that occupation cannot have started in Sveigakot much later than ca. A.D. 880 (Vésteinsson 2010). If Sveigakot was settled so early, it stands to reason that practically all other sites in the country must have been too (see Vésteinsson and McGovern 2012 for an elaboration of this argument). This assumption is supported by recent results from attempts to date other sites in Mývatnssveit. So far, archaeological deposits predating the ~A.D. 930–940 tephra have been found at six sites. Importantly, these include all site types, from un- or intermittently manned sites like shielings and small enclosures to small and middling farms and major estates. At two sites towards the upper end of the social scale, Hrísheimar and Þorleifsstaðir, archaeological deposits have been found directly on top of the landnám tephra, just like in Sveigakot, suggesting that Sveigakot was no anomaly regarding its early occupation. There are more than 20 other sites in Mývatnssveit with Viking age dates, but at only one, the feasting hall at Hofstaðir (Lucas 2009), are the features thus far investigated believed to date from later than ca. A.D. 930–940. Hofstaðir may yet yield earlier dates as its farm-mound is all but un-investigated and the midlate 10th-century building of the feasting hall likely represents the development of new political institu- Figure 1. Earliest phase of a pit house in Sveigakot built ca. 880 and abandoned well before ca. 940. Journal of the North Atlantic O. Vésteinsson and H. Gestsdóttir 2016 Special Volume 7 139 tions rather than pioneer farming (Friðriksson et al. 2004, Lucas 2009, Vésteinsson 2007). At the other Mývatn sites, the dating is either too general (e.g., based on artifact typology) or the dated features cannot be taken as representative of the site as a whole. Mývatnssveit is the most intensively studied region in Iceland, and comparable datasets do not exist for other parts of the country. The available evidence from other regions, such as Skagafjörður, Þjórsárdalur, and the Reykjavík area, is consistent with the interpretation suggested for Mývatnssveit. It would be difficult to argue that Mývatnssveit was a candidate for particularly early settlement. Although rich in freshwater fish and fowl, at 250 m above sea level it is less than ideally suited to the type of cattle-based farming that the settlers seem to have preferred (Vésteinsson 1998). In the same way as Sveigakot can be seen as a site of last resort and therefore a good indicator of when all land in that region had been occupied, Mývatnssveit as a whole can also be taken as an indicator of the same visa- vis the country as a whole. If Mývatnssveit was densely occupied before the end of the 9th century, so must most other parts of the country have been. An intensive survey and dating program in Langholt in Skagafjörður is underway and will hopefully throw clearer light on this issue (for early results interpreted differently than here, see Bolender et al. [2008, 2011]). Pollen data is consistent with a rapid and comprehensive occupation of the land in the decades around A.D. 900. This evidence is most graphically illustrated in a pollen study in Grímsnes in Southern Iceland where birch had fallen from a dominating position in the ecosystem to its present level before A.D. 920 (Hallsdóttir 1987:34, 1996:132), representing deforestation on a scale difficult to reconcile with anything but comprehensive occupation of that landscape. Of course, showing that there was extensive settlement before A.D. 920–940 does not reveal the whole story. It is important to note also that more than 100 sites have archaeological deposits on top of or just above the landnám tephra from A.D. 871±2. At two sites, there is evidence for human presence under this tephra. Both are in coastal locations in the Southwest: one in Krýsuvík and another in Reykjavík. In both cases, turf-walls, belonging to enclosures rather than houses, predate the tephra. It is, however, significant that neither at these sites, nor at any of the more than 100 sites where archaeological deposits have been observed in relation to the landnám tephra, have middens, burials, or dwellings—in other words evidence of a permanent settlement— been found under it. These findings suggest that people had begun to settle in Iceland before the deposition of the landnám tephra, but probably only shortly before and primarily in coastal areas, possibly only in the south. Shortly after the eruption, there was a massive immigration to Iceland which seems to have led to the establishment of large and small farms in practically all habitable parts of the country within a period of ~20 years. Based on an extrapolation from the Mývatn data, a recent estimate puts the minimum number of people needed to occupy all these posts at 24,000 (Vésteinsson and McGovern 2012). Taking into account that the first colonists are likely to have faced exaggerated mortality and that return migration may have occurred (King 1978), the actual number of people shipped to Iceland in these decades must have been considerably higher. But was that it? Was the migration to Iceland simply a somewhat faster process than Ari fróði imagined, 20 years instead of 60, with no significant additions after the initial land-rush? The results of strontium isotope analyses suggest that the story is not so simple, and in fact that it is much more complicated. Strontium Isotopes and the Nature of Colonization A recent study of strontium isotopes in human bones from Viking age Iceland has identified as immigrants at least 32 individuals out of a sample of 83 pagan burials in Iceland. In addition, one individual from a sample of 43 graves from two early Christian cemeteries turned out to be an immigrant (Price and Gestsdóttir 2006, this volume). The pagan burials are as a group dated to the late 9th and 10th centuries, whereas the Christian cemeteries are considered to belong to the 11th century and possibly later. The Christian cemetery at Skeljastaðir, where the immigrant comes from, is associated with radiocarbon determinations placing its use between the mid-10th and the early 13th centuries (Sveinbjörnsdóttir et al. 2010), so in that case there may be some overlap with the later pagan burials. Dating the pagan burials is not straightforward. Only three have a known association with a 10thcentury tephra layer, the E-934±2. All three are in Hrífunes in Southern Iceland (Kt 155)1 and predate the tephra: an adult native-born Icelander of uncertain sex is firmly below the tephra, an immigrant adult woman was buried at the time of the eruption (Larsen and Þórarinsson 1984:40–43), and an adult woman was buried and the burial subsequently disturbed prior to the eruption (Gestsdóttir et al. 2015). Those of the other graves that can be dated at all Journal of the North Atlantic O. Vésteinsson and H. Gestsdóttir 2016 Special Volume 7 140 have been dated either through radiocarbon determinations or artifact typologies, neither of which methods allow tight dating in this period. It is clear, however, that while many of the datable burials with immigrants may be from the late 9th century or early 10th century, a significant number seem to be later, from the second half of the 10th century. One of these late arrivals is an adult male (?) buried with a horse and a dog in Dalvík, northern Iceland, radiocarbon dated to ca. A.D. 978–1027 (Sveinbjörnsdóttir et al. 2010:686).2 Another immigrant in Dalvík is a woman aged 25–35 buried in one of the largest mounds recorded in Iceland (Fig. 2). She was accompanied by beads dated to the second half of the 10th century (Hreiðarsdóttir 2005[II]:105).3 In Sturluflötur, E-Iceland (Kt 136) an immigrant adult of uncertain sex was buried with a bead dated to A.D. 960–1000 (Hreiðarsdóttir 2005[II]:28). A highly unusual burial, if it is a burial at all, is that of an adult found on top of a mountain in eastern Iceland. Accompanied by no fewer than four brooches and slightly less than 500 beads, this individual may have sought shelter in a cave and died there, rather than being intentionally buried in this atypical location. Whatever the case, this individual was an immigrant whose most recent beads were made after A.D. 960 (Hreiðarsdóttir 2005[II]:111). Including the immigrant in the Christian cemetery at Skeljastaðir, five out of 32 identified immigrants arrived in Iceland in the late 10th century or later. A further two immigrants from pagan grave contexts (in Hafurbjarnarstaðir [Kt 40] and Smyrlaberg [Kt 65]) have anomalously late radiocarbon dates (11th–13th century) that will need corroboration. Discounting these, five out of 30 is a significant number, considering that only seven of the immigrants have firm late 9th and early 10th century dates, while 11 have general 10th-century dates (i.e., unlikely to have arrived in the late 9th century) and seven have only general Viking age dates. These dating issues would repay closer examination than has been attempted here, but the implications are clear: people continued to arrive in Iceland throughout the 10th century and possibly longer. On the face of it, this evidence would seem to be more consistent with Ólafur Lárusson’s model of gradual settlement increase, but how can this be reconciled with the indications for very early and comprehensive settlement in Mývatnssveit? The answer is simple enough: While very large numbers of people were shipped to Iceland in the late 9th century, enough to occupy the same, or larger, number of settlements as occupied in post-Viking age times, immigration also continued for a long time afterwards. This scenario has several implications. It is possible that despite large numbers, the Figure 2. One of the burial mounds in Dalvík under excavation by Daniel Bruun and Finnur Jónsson in 1909. © Nationalmuseet. Journal of the North Atlantic O. Vésteinsson and H. Gestsdóttir 2016 Special Volume 7 141 late 9th century immigrant population was not able to adequately sustain itself, let alone increase, so that reinforcements were needed for a long time. It may also be that despite large numbers of immigrants in the late 9th century and a possibly robust natural increase of that population, there was still a need for more immigrants and/or continued pressures to migrate to Iceland. Another implication is that the postulated late-9th-century population is severely underrepresented in the burial data. These can be seen as competing assertions, but we suggest they are complementary: all are true. The possible inability of the late-9th-century population to sustain itself could have been caused by three factors (at least). Exaggerated mortality, through hardship, apathy, or violence is frequently observed in founding populations (e.g., Kupperman 1979) but is usually only associated with the very first years of settlement. Modern migration studies show that return-migration is an aspect of all migrations (Gmelch 1980, King 1978, Wyman 1993) and it is likely that this was so in prehistory too, although it is notoriously difficult to detect. Studies of pre-steamship return migration from America suggest that it is likely to have been less than 1% of the out-migration (Kamphoefner 1991:297–301). A third possible factor is an unbalanced sex ratio. If men significantly outnumbered women—as they frequently do in frontier situations (Sharpe 2001, Simon and Brettel 1986)—then natural fertility, even if exaggerated as it often is in new societies (Harris 2001:13–143), may not have been enough to sustain the population. It is possible that such factors were significant enough to require a steady influx of people for more than a century after the initial land-rush, but their impact may equally have been felt primarily in the very first years of the colonization, perhaps in the pre-A.D. 870 period, and robust fertility rates might have begun to assert themselves as early as the A.D. 880s or 890s. The strontium evidence allows us to suggest a more nuanced scenario. Patterns in Burial Data It might be assumed that the roughly 320 pagan burials known in Iceland reflect the population as a whole. Alas, it is not so simple, and it is well known that this sample is biased in a number of ways. The very low incidence of children is one obvious thing to note, with only two infants and 10 children aged 7–17 out of a population of 119 preserved skeletons (Friðriksson 2000:594). Another is the skewed sex ratio, with 35 women against 73 men (Friðriksson 2000:595). Neither aspect is unique for Iceland, but they do suggest that significant proportions of the population are missing from the burial record. This implication underpins general suspicions, also entertained, e.g., in Norway (Nordeide 2011, Solberg 2000:268), that the burial record represents primarily the upper portions of the social scale. One way of approaching this is to look at the size of pagan cemeteries. The majority of pagan burials are isolated finds, and only a handful of cemeteries have been examined comprehensively. The largest has 12 human graves, but Adolf Friðriksson (2009:12–14) concludes that a typical Icelandic grave field has at least 4–5 graves.4 This is probably an over-cautious estimate and the existence of cemeteries with 9–12 graves suggests that this could have been the norm. Even so, this is much too few graves to account for all the people who must have lived on the farms with which these cemeteries are associated. If the pagan cemeteries each only served one farm (and this is not certain) and we reckon with a very low average household size , say 5 compared to 6.15 in the 18th century (Jónsson and Magnússon 1997:140), then with a “normal’” mortality rate of 30‰ (Jónsson and Magnússon 1997:188) and a century of use, each cemetery should have at least 17 human burials. This is an absolute minimum because the period of use is likely to have been longer, the mortality rate higher, especially to begin with, and the household size larger, making an expected number of dead per farm in the pagan period closer to 20–25. However we juggle these figures, it is clear that the pagan cemeteries are missing at least a third and more likely two thirds or more of the actual population. We know that infants in particular, but also children and women, make up a significant proportion of those missing. However, adding likely figures for those would not make up the shortfall. In other words, some adult males are also missing, and the suspicion is that these are the dispossessed and unconnected: slaves, servants, and others of low status however defined. The burial record as a whole therefore represents only a specific proportion of Icelandic society in the 9th and 10th centuries. Looking at the immigrants within this biased sample is quite revealing. Two things stand out: Firstly, unlike the assemblage as a whole, the sex ratio is nearly even among the immigrants. There are 11 females or likely females against 12 males or likely males compared to a ratio of 31:10 males to females among those born in Iceland. In other words, a disproportionally high ratio of the women buried in these graves are immigrants. The second is that the immigrants are associated with richer burials. This can be demonstrated in two ways. First, native born Icelanders are much more frequently found in graves with no grave-goods (16 out of 51 Journal of the North Atlantic O. Vésteinsson and H. Gestsdóttir 2016 Special Volume 7 142 In part, this situation probably relates to affluence; the bulk of the founding population was likely very poor and only later generations had the wherewithal to invest in objects with symbolic meaning. But it also relates to levels of social differentiation and the need to assert status and social inclusion. We could imagine that the founding population was largely unfree, and uniformly so, governed perhaps by overseers and/or a very small group of landowners, many of whom may not have permanently settled in Iceland. That would fit with numerous examples from early modern colonizations where the people who invested in colonizing ventures, and expected to increase their wealth and power from this, did not participate in person, or only fleetingly so. Instead people of low or questionable status were sent out to do the actual colonizing only to be followed later by more respectable types or by developing respectability themselves in later generations (for variations in such developments in possibly analogous situations see Choquette [1997], Pope [2004]). In contrast to the invisibility of the first one or two generations in Iceland, whose dwelling remains left no positive features and whose burials only rarely had the grave goods or other features which would help us find them, their descendants in the late 10th century employed massive turf architecture and buried their dead much more frequently with grave goods. As a result, we are much better informed about the latter, and we tend to assume that the evidence from their remains holds for the Icelandic Viking age as a whole. This assumption is not valid, and making the distinction holds the key to understanding the complex goings-on in this formative society. Becoming Respectable in 10th-Century Iceland Of the 16 burials defined by Kristján Eldjárn and Adolf Friðriksson as the richest in the country, none can be dated to the late 9th century, only two to the early 10th century,8 while at least six belong to the mid- to late 10th century.9 We can take these findings as a rough indication of the incidence of respectability in Viking age Iceland. It was low to begin with and possibly associated mainly with people who did not settle permanently and are therefore not present in the Icelandic burial record, but increased as the 10th century wore on. An important stage will have been reached when an indigenous upper-class had emerged, which may plausibly be associated with the development between A.D. 930 and the 960s (Ari’s dates) of a hierarchical system of assemblies applying local law to the settlement of disputes. This native upper class will have wanted to wrest any dicases [31%]), but only two out of 32 immigrants (6%) come from graves with no grave goods. Second, of the 16 burials characterized by Kristján Eldjárn and Adolf Friðriksson as the richest in the country (Eldjárn 2000:303-304), six are immigrants5 while only two are native-born Icelanders6 (the remaining eight either have no bones preserved or the teeth have not been analysed for strontium isotope ratios). These findings can be interpreted in a variety of ways, but we propose that these patterns all relate to the same phenomenon, namely the double invisibility of the late 9th-century founding population and the exaggerated visibility of an upper class with partly foreign roots and a definite Norse identity emerging in the late 10th century. The Dispossessed in Early Iceland The people who established more than 30 farms in Mývatnssveit, and by implication more than 4000 in the country as a whole, in the late 9th century are doubly invisible in the archaeological record. There are too few graves from this period (even if we think that they were all young people who lived long lives—which is unlikely). The simplest explanation is that this dearth is caused by the same factor as barred a significant proportion of the population from burial in cemeteries throughout the pagan period: low status and absence of family ties and obligations. The other aspect of this invisibility is more concrete: in Mývatnssveit, practically all positive features which have been dated in relation to the ~A.D. 930–940 tephra post-date it. All home-field enclosures and field boundaries are later and so are all the halls with the characteristic plan of Norse dwellings in the Viking age (bow-shaped walls, three aisled, central hearth, etc.).7 So far pre- A.D. 930–940 dwellings have only been revealed at Sveigakot, and these are sunken-featured buildings without any particularly Norse ethnic markers (Vésteinsson 2010). Apart from that, the pre-A.D. 930–940 cultural deposits consist of middens, upcast (presumably from sunken-featured buildings), and traces of cultivation/field improvements. The artifacts from these early layers are both limited in volume and singularly non-specific in their associations. There is nothing in these assemblages suggesting Norse, or indeed any other specific ethnicity. This evidence is in contrast to the late-10th-century assemblages, which are both larger in volume and diversity and also have artifacts with symbolic meaning. In Sveigakot, these include a piece from an oval brooch and a dragon head from a bone pin from layers contemporary with a very small, but also very Norse, hall (Vésteinsson 2014). Journal of the North Atlantic O. Vésteinsson and H. Gestsdóttir 2016 Special Volume 7 143 Eldjárn, K. 2000. Kuml og haugfé úr heiðnum sið á Íslandi. 2nd Edition. Adolf Friðriksson (Ed.) Mál og menning, Reykjavík, Iceland. Friðriksson, A. 2000. Viking burial practices in Iceland. Pp. 549–610, In A. Friðriksson (Ed.). Kuml og haugfé úr heiðnum sið á Íslandi. 2nd Edition. Mál og menning, Reykjavík, Iceland. Friðriksson, A. 2009. Social and symbolic landscapes in Late Iron Age Iceland. Archaeologia islandica 7:9–21. Friðriksson, A., and O. Vésteinsson. 2003. Creating a past: A historiography of the settlement of Iceland. Pp. 139–161, In J. Barrett (Ed.). Contact, Continuity, and Collapse: The Norse Colonization of the North Atlantic (Studies in the Early Middle Ages 5). Brepols, Turnhout, Belgium. Friðriksson, A., O. Vésteinsson, and T.H. McGovern. 2004. Recent investigations at Hofstaðir, northern Iceland. Pp. 191–202, In R.A. Housley and G. Coles (Eds.). Atlantic Connections and Adaptations: Economies, Environments, and Subsistence in Lands Bordering the North Atlantic. Oxbow Books, Oxford, UK. Gestsdóttir, H., U. Ævarsson, G.A. Gísladóttir, and E.Ó. Hreiðarsdóttir. 2015. Kumlateigur í Hrífunesi í Skaftártungu V. Árbók Hins íslenzka fornleifafélags 2014:7–34. Gmelch, G. 1980. Return migration. Annual Review of Anthropology 9:135–159. Grönvold, K., N. Óskarsson, S.J. Johnsen, H. Clausen, C.U. Hammer, G. Bond, and E. Bard. 1995. Ash layers from Iceland in the Greenland GRIP ice core correlated with oceanic and ash sediments. Earth and Planetary Science Letters 135:149–155. Hallsdóttir, M. 1987. Pollen analytical studies of human influence on vegetation in relation to the landnám tephra layer in Southwest Iceland. Lundqua Thesis 18:1–45. Hallsdóttir, M. 1996, Frjógreining. Frjókorn sem heimild um landnámið. Pp. 123–134, In G.Á. Grímsdóttir (Ed.). Um landnám á Íslandi: Fjórtán erindi. Vísindafélag Íslendinga, Ráðstefnurit, Reykjavík, Iceland. Harris, P.M.G. 2001. The History of Human Populations. Vol. 1: Forms of Growth and Decline. Prager, Westport, CT, USA. Hermanns-Auðardóttir, M. 1989. Islands tidiga bosättning. Studier med utgångspunkt i merovingertida-vikingatida gårdslämningar i Herjólfsdalur, Vestmannaeyjar, Island. Studia archaeologica Universitatis Umensis, I, Umeå, Sweden. Hermanns-Auðardóttir, M. 1991. The early settlement of Iceland: Results based on excavations of a Merovingian and Viking farm site at Herjólfsdalur in the Westman Islands, Iceland. Norwegian Archaeological Review 24:1–9. Hreiðarsdóttir, E.Ó. 2005. Íslenskar perlur frá víkingaöld – með viðauka um perlur frá síðari öldum. 2 vols. Unpubl. M.A. Thesis. University of Iceland, Reykjavík, Iceland. Jónsson, G., and M.S. Magnússon (Eds.). 1997. Hagskinna: Sögulegar hagtölur um Ísland. Haǵstofa Íslands, Reykjavík, Iceland. rect influence over Icelandic affairs and possessions from foreign competitors (if the latter had not turned to other more lucrative ventures already), but they will also have wanted to increase their respectability through both direct and symbolic associations with established aristocracies in the perceived homelands. It is clear that, whatever the ethnic origins of the 9th-century population, by the mid- to late 10th century the source of symbolic power for these Icelandic wannabes was Scandinavia (Vésteinsson 2014). It is in this light that we need to consider the continued immigration to Iceland in the second half of the 10th century. There may still have been a need for workers and tenants, but the relatively greater number of wealthy people and women among the immigrants suggests that there had developed demand for people of a different sort. In particular, it is easy to see how respectability could be sought through marriage with Scandinavian aristocrats, or even just any free-born Scandinavian (see Metcalf [2005] for an example of a colonial aristocracy maintaining status through marriages with impecunious noblemen from the homeland for more than two centuries after settlement). Perhaps the mound erected over the young immigrant woman in Dalvík around 1000, one of the largest known in Iceland (Fig. 2), represents an attempt to advertise the respectability which had almost been secured, but was thwarted by an untimely death? Acknowledgments The authors are grateful to T. Douglas Price for organizing the workshop in Copenhagen in 2011 where the results of this research were presented. This research forms part of two Rannís-funded projects: “Death and Burial in Iceland for 1150 Years” (grant no. 110646021) and “The Settlement of Iceland: The Analysis of Strontium Isotopes in Human Remains” (grant no. 40636031). Literature Cited Bolender, D.J., J.M. Steinberg, and E.P. Durrenberger. 2008. Unsettled landscapes: Settlement patterns and the development of social inequality in northern Iceland. Pp. 217–238, In C.A. Pool and L. Cliggett (Eds.). Economies and the Transformation of Landscape. AltaMira Press, Lanham, MD, USA. Bolender, D.J., J.M. Steinberg, and B.N. Damiata. 2011. Farmstead reorganization at the end of the Viking Age. Results of the Skagafjörður archaeological settlement survey. Archaeologia Islandica 9:77–101. Choquette, L. 1997. Frenchmen into Peasants. Modernity and Tradition in the Peopling of French Canada. Harvard University Press, Cambridge, MA, USA. Eldjárn, K. 1956. Kuml og haugfé úr heiðnum sið á Íslandi. Norðri, Akureyri, Iceland. Journal of the North Atlantic O. Vésteinsson and H. Gestsdóttir 2016 Special Volume 7 144 Kamphoefner, W.D. 1991. The volume and composition of German-American return migration. Pp. 293–311, In R. Vecoli and S.M. Sinke (Eds.). A Century of European Migrations, 1830–1930. University of Illinois Press, Urbana, IL, USA. King, R. 1978. Return migration. A neglected aspect of population geography. Area 10:175–182. Kupperman, K.O. 1979. Apathy and death in early Jamestown. Journal of American History 66:24–40. Larsen, G., and S. Þórarinsson. 1984. Kumlateigur í Hrífunesi í Skaftártungu IV. Árbók hins íslenzka fornleifafélags 1983:31–47. Lárusson, Ó. 1929. Úr byggðarsögu Íslands. Vaka 3:319–369. Lucas, G. 2009. Hofstaðir. Excavations of a Viking Age Feasting Hall in Northeastern Iceland. Institute of Archaeology, Monograph Series 1, Reykjavík, Iceland. McGovern, T.H., O. Vésteinsson, A. Friðriksson, M. Church, I. Lawson, I. A. Simpson, Á. Einarsson, A. Dugmore, G. Cook, S. Perdikaris, K.J. Edwards, A.M. Thomson, W.P. Adderley, A. Newton, G. Lucas, R. Edvardsson, O. Aldred, and E. Dunbar. 2007. Landscapes of Settlement in Northern Iceland: Historical Ecology of Human Impact and Climate Fluctuation on the Millennial Scale. American Anthropologist 109:27–51. Metcalf, A.C. 2005. Family and Frontier in Colonial Brazil: Santana de Parnaíba, 1580–1822, 2nd Edition. University of Texas Press, Austin, TX, USA. Nordeide, S.W. 2011. Where did all the people go? Looking for eleventh-century graves in southern Norway. Pp. 320–332, In S. Sigmundsson (Ed.) Viking Settlements and Viking Society. Papers from the Proceedings of the Sixteenth Viking Congress, Reykjavík and Reykholt, 16–23 August 2009. University of Iceland Press, Reykjavík, Iceland. Pope, P.E. 2004. Fish into Wine: The Newfoundland Plantation in the Seventeenth Century. University of North Carolina Press, Chapel Hill, NC, USA. Price, T.D., and H. Gestsdóttir. 2006. The first settlers of Iceland: An isotopic approach to colonisation. Antiquity 80:130–144. Price, T.D., and H. Gestsdóttir. In review. The peopling of the North Atlantic: Isotopic results from Iceland. Journal of the North Atlantic. Sharpe, P. (Ed.). 2001. Women, Gender, and Labour Migration: Historical and Global Perspectives. Routledge, London, UK. Sigurgeirsson, M.Á., U. Hauptfleisch, A. Newton, and Á. Einarsson. 2013. Dating of the Viking Age Landnám tephra sequence in Lake Mývatn sediment, North Iceland. Journal of the North Atlantic 21:1–11. Simon, R.J., and C.B. Brettel (Eds.). 1986. International Migration: The Female Experience. Rowman and Allanheld, Totowa, NJ. Solberg, B. 2000. Jernalderen i Norge. ca. 500 f.Kr.–1030 e.Kr. Cappelen akademisk forlag, Oslo, Norway. Sveinbjörnsdóttir, Á.E., J. Heinemeier, J. Arneborg, N. Lynnerup, G. Ólafsson, and Z. Guðný. 2010. Dietary reconstruction and reservoir correction of 14C dates on bones from pagan and early Christian graves in Iceland. Radiocarbon 52(2–3):682–696. Thomson, A.M., and I.A. Simpson. 2007. Modeling historic rangeland management and grazing pressures in landscapes of settlement. Human Ecology 35:151–168. Vésteinsson, O. 1998. Patterns of settlement in Iceland: A study in prehistory. Saga-Book of the Viking Society 25:1–29. Vésteinsson, O. (Ed.). 2004. Fornleifarannsóknir í Saltvík 2003, Reykjavík, Iceland. Vésteinsson, O. 2007. “Hann reisti hof mikið hundrað fóta langt ...” Um uppruna hof- örnefna og stjórnmál á Íslandi í lok 10. aldar. Saga. Tímarit Sögufélags 45:53–91. Vésteinsson, O. 2010. Ethnicity and class in settlement period Iceland. Pp. 494–510, In John Sheehan and Donnachadh Ó Corráin (Eds.). The Viking Age: Ireland and the West. Papers from the Proceedings of the Fifteenthe Viking Congress, Cork, 18–27 August 2005. Four Courts Press, Dublin, Ireland. Vésteinsson, O. (Ed.). 2011. Archaeological investigations in Mývatnssveit, Reykjadalur and Svartárkot 2010. Fornleifastofnun Íslands, Reykjavík, Iceland. Vésteinsson, O. 2012. Gröf og dauði. Skýrsla um vettvangsrannsóknir sumarið 2011. Fornleifastofnun Íslands, Reykjavík, Iceland. Vésteinsson, O. 2014. Shopping for identities: Norse and Christian in the Viking Age North Atlantic. Pp. 75–91, In Ildar Garipzanov, with assistance of Rosalind Bonté (Eds.). Conversion and Identity in the Viking Age. Brepols, Turnhout, Belgium. Vésteinsson, O., and T.H. McGovern. 2012. The peopling of Iceland. Norwegian Archaeological Review 45(2):206–218. Vésteinsson, O., T.H. McGovern, and C. Keller. 2002. Enduring Impacts. Social and Environmental Aspects of Viking Age Settlement in Iceland and Greenland. Archaeologia islandica 2:98–136. Vilhjálmsson, V.Ö. 1989. Stöng og Þjórsárdalur‑bosættelsens ophør. Hikuin 15:75–102. Wyman, M. 1993. Round-trip to America. The Immigrants Return to Europe, 1880–1930. Cornell University Press, Ithaca, NY, USA. Zielinski, G.A., M.S. Germani, G. Larsen, M.G.L. Baille, S. Whitlow, M.S. Twickler, and K.C. Taylor. 1997. Volcanic aerosol records and tephrochronology of the Summit, Greenland, ice cores. Journal of Geophysical Research 102:26,625–26,640. Endnotes 1“Kt” followed by a number refers to the numbering of pagan burial sites in Eldjárn (2000) and Friðriksson (2000). 2DAV-A-009 = kuml 12. Kt. 89. The skeleton is wrongly described as female in Sveinbjörnsdóttir et al. (2010:686). 3DAV-A-008 = kuml 13. Kt 89 (Hreiðarsdóttir 2005[II]:105). Journal of the North Atlantic O. Vésteinsson and H. Gestsdóttir 2016 Special Volume 7 145 4In fact, in his paper, Adolf Friðriksson published maps of robbed and unexcavated cemeteries with 13 (Berufjörður) and 17 graves (Ingiríðarstaðir). Some of these are likely to be horse graves, but this evidence supports the case for substantially larger cemeteries than 4–5. 5Hafurbjarnarstaðir (Kt 40); Sílastaðir (Kt 98) graves 1, 2, and 4; Daðastaðir (Kt 126); and Álaugarey (Kt 151). Of these, the association of bones with grave-goods in the Hafurbjarnarstaðir burial is uncertain. 6Öndverðarnes (Kt 47) and Vatnsdalur (Kt 54). 7This is true of the fully excavated halls at Hofstaðir (Lucas 2009) and Sveigakot, as well as hall-like structures trenched in Girðingar, við Kleifarhólma (Vésteinsson 2011:27, 35), Raufarhóll (Vésteinsson 2012:12), and Saltvík (Vésteinsson 2004:11–13). To this can be added evidence from other parts of the country, e.g., the earlier hall in Stöng, which post-dates the E-934 ± 2 tephra (Vilhjálmsson 1989). 8 Daðastaðir (Kt 126) and Öndverðarnes (Kt 47)—the latter judging only from its M-type sword. 9Miklaholt (Kt 36), Eyrarteigur (Kt 144), Vatnsdalur (Kt 54), Kornsá (Kt 63), and Ketilsstaðir (Kt 142) are dated to the mid-10th century, while Baldursheimur (Kt 117) only has a general-10th-century date. A comparable picture emerges when burials with datable beads are arranged chronologically: only one has a 9th-century date and two have firm early-10th-century dates, while 14 have firm late-10th-century dates and the remaining 4 could be from the late 9th to the late 10th (Hreiðarsdóttir 2005[I]:165).