Eagle Hill Masthead

Journal of the North Altantic
    JONA Home
    Aim and Scope
    Board of Editors
    Editorial Workflow
    Publication Charges

Other Eagle Hill Journals
    Northeastern Naturalist
    Southeastern Naturalist
    Caribbean Naturalist
    Neotropical Naturalist
    Urban Naturalist
    Prairie Naturalist
    Eastern Paleontologist

Eagle Hill Institute Home

About Journal of the North Atlantic


Modelling Viking Migration to the Inner Hebrides
Alan Macniven

Journal of the North Atlantic, Special Volume 4 (2013): 3–18

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


Site by Bennett Web & Design Co.
A. Macniven 2013 Journal of the North Atlantic Special Volume 4 3 Introduction When it comes to the Viking Age (ca. A.D. 800– 1000),1 the islands to the west of Scotland are not well served by the documentary record. Contemporary accounts of Scandinavian activity in the area are not only limited but very highly polarized. With the emphasis throughout being on violence and aggression rather than diplomacy or domestic affairs (e.g., AU 794.7, 798.2; ASB 847),2 and the only Hebridean target specified by name being the wealthy monastery of Columba on Iona (AI 795.2; AU 802.9, 806.8, 825.17, 878.9, 986.3),3 it is easy to see why traditional explanatory models have focussed on Victorian notions of piracy, plunder, and seasonal exploitation (e.g., Smyth 1984:141–174).4 If the investigation is expanded to consider other types of evidence, however, it is clear that the Scandinavian presence in the Hebrides soon took on a more permanent character. The combined witness of many hundreds of locally preserved Scandinavian artifacts (Graham-Campbell and Batey 1998, Shetelig 1954), place-names (Cox 2002; Johnston 1991; Macniven, in press; Oftedal 1954; Stahl 1999; Thomas 1874–1876, 1881–1882), loanwords (Gammeltoft 2004, Henderson 1910, Stewart 2004), and linguistic features (Borgstrøm 1974, Gameltoft 2004, Oftedal 1961) suggests, moreover, that the Age of “Viking” settlement was one of cultural change. Observations on the nature of that change have varied considerably. Whereas studies of the Outer Isles have tended to converge on the violent displacement of established populations by ethnic Scandinavians (cf. Jennings and Kruse 2005, 2009a, 2009b), surveys of the Inner Isles have generally placed greater emphasis on ethnic continuity. Even where the possibility of Norse migration is accepted, there are usually caveats regarding the scale, localization, or speed with which the immigrants were absorbed into pre-existing cultural groups (e.g., Heather 2009:488). The result has been the development of a conceptual “North versus South” divide, which has grown into something of a self-referencing dogma. This, in turn, has led to an imbalance in the selection and treatment of evidence, clouding our understanding of the protagonists’ motivations, and serving mainly to stifle debate. One way to move the discussion forward is to explore the possibility of a more uniform narrative for the Hebrides as a whole. Recent developments in “migration theory” are instructive here—and in particular the concept of “predatory migration” as developed by Heather (2009) and Halsall (2007)— but these can be further refined to reflect local circumstances. In so doing, it makes sense to follow the lead of Ó Corráin (1998b), Jennings and Kruse (2005:293), Woolf (2007:286–300), and others in approaching the evidence in terms of formative push and pull factors. The present paper will attempt to do so with a focus on the Inner Hebridean island of Islay (Fig. 1). Lying at the center of the early medieval Gàidhealtachd, Islay might seem an unlikely candidate for Scandinavian invasion—but it is for that very reason the ideal test bed for such a theory. Continuity or Change? The idea of a North versus South divide in Hebridean Viking studies has its roots in the perceived contrast in cultural continuity around those poles from the early Christian Era to more recent times. With the Outer Isles believed to have been Pictish before the Viking Age but Gaelic afterwards (Kruse 2005:148–151, Jennings and Kruse 2009a:75–79), there is certainly scope for considering Norse settlement as the main agent of cultural change. It has been argued by Jennings and Kruse (2005:288–292), for example, that the post-medieval dominance of the Gaelic language in the area may even have begun Modelling Viking Migration to the Inner Hebrides Alan Macniven* Abstract - Until recently, the idea of large-scale Viking settlement in Scotland’s Inner Hebrides was considered unlikely. Despite a conspicuous absence of documentary evidence, the area’s long-standing Gaelic heritage was seen as proof of linguistic and cultural continuity from its Dalriadan heyday. By developing the narrative to consider other types of evidence, however, it is clear that the Norse impact on these islands was far from insignificant. This paper will review the historical record in the light of material evidence and linguistic artifacts such as place-names. After questioning aspects of currently popular approaches to “predatory” migration, it will then examine how reappraisal of the practicalities of Viking Age immigration might help to inform a revised model for Norse settlement in the region. Across the Sólundarhaf: Connections between Scotland and the Nordic World Selected Papers from the Inaugural St. Magnus Conference 2011 Journal of the North Atlantic *Division of European Languages and Cultures: Scandinavian Studies, The University of Edinburgh, 13.05 David Hume Tower, George Square, Edinburgh EH8 9JX Scotland; Alan.Macniven@ed.ac.uk. 2013 Special Volume 4:3–18 A. Macniven 2013 Journal of the North Atlantic Special Volume 4 4 with the Norse importation of craftsmen and slaves from Ireland. Further south, on the other hand, the situation appears to have been quite different. Take, for example, the isle of Islay. In the eighth century A.D., on the eve of the Viking Age, Islay was home to the powerful kindred of Oengus, and as such fully integrated into the Gaelic-speaking kingdom of Dàl Riata (Bannerman 1974). There are no contemporary accounts of Viking activity on the island,5 or of any other events between A.D. 740 and 1095,6 but as the Gaelic language also appears to have defined local identity under the MacSorley Kings of the Isles from at least the middle of the twelfth century (Caldwell 2008:33–48), presumptions have been of continuity rather than change, with any lasting Norse influence assumed to have been minimal or the result of gradual accretion (cf. Barrett 2008b:413, Marsden 2000:12–13, Mc- Donald 1997:28, Nieke 1983:313, Storrie 1997:32). Arguments for continuity have been based on three main areas of evidence: language, place-name ratios, and a lack of convincing material evidence for settlement. While superficially convincing, these arguments tend to rely on the simple restatement of statistics without typological analysis of the available evidence or due consideration of how the underlying datasets have been shaped. Reviewing the Evidence Linguistic continuity With the isles to the west of Scotland flanking the main medieval transit route from Norway to the Irish Sea, it can be assumed that the Inner Hebrides witnessed a high volume of Scandinavian sea-traffic during the Viking Age (cf. Crawford 1987:19). As such, there is a reasonable possibility that at least some of the traces of the Norse language which have survived in situ in Islay and the surrounding area were introduced directly by native speakers, rather than indirectly through the migration of ideas. Given that the post-Viking Age language of the area was Gaelic and not Norse,7 it might nevertheless be imagined that any actual Norse settlement was limited in scale or lacking the social standing needed to impact on the language of the dominant speech community. Gammeltoft’s (2004:63) recent survey of MacBain’s Etymological Dictionary of the Gaelic Language suggests that Norse loanwords account for less than 3% of the total, less than a third of the corresponding figure for English and half the figure for Latin (cf. Oftedal 1961:120). At the same time, it is pertinent to note that Scottish Gaelic has diverged noticeably from that of Ireland since the beginning of the Viking Age. Closer scrutiny of Figure 1. Location of Islay. A. Macniven 2013 Journal of the North Atlantic Special Volume 4 5 those changes suggests that the most likely explanation is intimate contact with Old Norse. According to sociolinguist Uriel Weinreich (1974:2), there are three main ways in which languages in contact influence each other: grammatically, lexically, and phonetically. In terms of grammar, there is little direct evidence for Norse influence on Scottish Gaelic. While it does appear to have undergone the kind of general morphological simplification during the Middle Ages that might be expected in language contact situations (cf. Gammeltoft 2004:59–61, Jackson 1951), this could also be explained by the all-pervading expansion of English. When it comes to lexis and phonetics, however, the indications are rather less ambiguous. The number of nouns borrowed from Old Norse into Gaelic, for example, may be limited in absolute terms, but vastly outweighs that borrowed from Gaelic into Scandinavian (Shulze-Thulin 2001:53), pointing to an extreme imbalance in the social status of the two languages. Norse loans into Gaelic tend, moreover, to be of the stem variety. Whether this was to ease their retention through the requirements of Gaelic grammar (Gammeltoft 2004:62–63) is open to question; but it is surely significant that a disproportionately high number begin in combinations of the letter /s/ + [stop]—exactly what we might expect if Norsespeakers were struggling with the Gaelic grammatical phenomenon of word-initial lenition (Stewart 2004:405–406). This linguistic evolution can be seen as an example of what Thomason (1987:181) calls “language-shift induced interference”, with the changes here resulting from the imperfect second language acquisition of Gaelic by native speakers of Old Norse rather than a borrowing of Scandinavian word material by a Gaelic language speech community. The development of phonetic features such as the devoicing of voiced stops, the introduction of initial stress, supradentalization, retroflexion, and the pre-aspiration of unvoiced plosives adds to this picture (Gammeltoft 2004:55–59). Taken as a whole, therefore, the linguistic evidence hints at a break in the Gaelic tradition of the isles, with the post-medieval variety representing a language which has been learned by native-speakers of Old Norse (Gammeltoft 2004:74, Stewart 2004:402–406) and suggesting a phase of ethnic disjuncture. Perceptions of ethnicity With language being a key component of ethnic identity in Early Medieval Britain (Jennings 1996:68, Woolf 2007:1–14), it is important to note that the name by which the Hebrides were known to their immediate neighbors also changed during the Viking Age. The designation “Hebrides”, recorded by Ptolemy ca. A.D. 120 as Ebudae, is rendered Ibd(a)ig in the Irish annals of the pre-Norse period (AU 557.1, 672.2). By the 10th century, however, these islands had become known as Innse Gall—the “Isles of the Foreigners”—but more specifically, the “Isles of the Scandinavians”, in contradistinction to the adjacent mainland of Argyll, earlier Airer Goidel, or the “Coastline of the Gael” (Sellar 1966:135; Woolf 2004:94–99; Woolf 2007:64, 100). Interestingly, the significance of this onomastic shift in marking a change in local ethnicity has been marginalized over the years by the circular restatement of a number of anecdotal accounts of cultural and linguistic blending. Principal among these is the series of five annalistic references to the Irish campaigns of a mysterious military faction known as the Gall Ghàidheil (FA 856 (§247), 858 (§260), 858 (§263); AU 856.3, 856.5). Although these “Gaelicspeaking Vikings” were previously assumed to have originated in the Hebrides (e.g., Smyth 1984:157), it seems unlikely they could reflect the established population of Innse Gall, certainly not after the mid- 9th century. To be a warrior in this period was to be a member of the nobility (cf. Halsall 1997). If resident Gaelic-speakers had been able to retain this kind of status through the early stages of the Viking Age, we might expect it to be mirrored in the linguistic evidence—which it is not. As a consequence, more recent studies of the group have sought to trace them to parts of the mainland or the Clyde estuary, where the terms of contact between Norse and native appear to have been different (cf. Clancy 2008, Jennings and Kruse 2009b). Attention has also been drawn to the Irish names and by-names of certain otherwise Norse characters in the later medieval Icelandic saga literature (cf. Jennings 1996, Jennings and Kruse 2009b). The originally 12th-century work known as Landnámabók (ON, The Book of Settlements), for example, lists around 400 main settlers and several thousand of their dependents thought to have comprised the initial wave of settlement in Iceland in the second half of the 9th century (Pálsson and Edwards 1972). With some of the leading families in this movement said to have come from the Suðreyjar, or Hebrides, been headed by "Irishmen", or to have brought with them a penchant for Gaelic fashions in personal names, such as Kalman (G Colmán) (Schulze-Thulin 2001:74), by-names, such as Helgi bjólan (G beulan, little mouth) (Schulze-Thulin 2001:70), “Irish” wives and slaves (Sigurdsson 1988:28), and even “Celtic” Christianity (cf. Guðmundsson 1997:101–120, Pálsson and Edwards 1972:23 FN 20), it has been assumed that these episodes must signify early cultural blending in the Isles (cf. Smyth 1984:162–163). As Friðriksson and Vésteinsson (2003) point out, however, it A. Macniven 2013 Journal of the North Atlantic Special Volume 4 6 is difficult to ascertain how much of the narrative here is historically accurate, and how much should be seen as an agenda-driven literary construct. It should, in any case, also be stressed that the earliest reliable accounts of settlement and society in Iceland point to a culturally united community with pagan Scandinavian values and a linguistic foundation in Old Norse. If the later use of the term “Irishmen” does, in fact, derive from an actual ethno-linguistic distinction, and not simply a Norse fashion, these individuals and their households must have integrated very quickly into Iceland’s prevailing cultural norms (cf. Schulze-Thulin 2001:53–54)—if they had not already done so in Ireland or the Hebrides. In Chapter 20 of Landnámabók, for example, we learn of a certain Avang, nominally “the Irishman”, who was “the father of Thorleif, father of Thurid, wife of Thormod” (Pálsson and Edwards 1972:25). Given that Avang’s descendents bear clearly Scandinavian names, it seems unlikely that he was a hardline champion of Gaelic tradition. If we accept that the use of Gaelic names and by-names in the Icelandic literary material is a genuine reflection of 9th-century tradition, it is also worth considering the “prestige deficit” incumbent on the indigenous population of the Hebrides by this time. No obituaries are recorded for kings of Dàl Riata after that of Donn Corci in AU 792.4— hinting at Viking agency in the downfall of the institution. If the remainder of the nobility were similarly depleted, there would have been little incentive for status-obsessed Norse migrants to emulate the local culture. It may be significant in this respect that saga episodes set in the Hebrides make no reference to the community at large being Gaelic- speaking. Indeed, it could be argued that the ease with which they seem to communicate with (presumably monoglot) Norse-speakers from Iceland or Norway (Power 1990) preserves the folk-memory of cultural disjuncture in the Isles. Considering the well-rehearsed exploits of Viking Age Scandinavians in Ireland, it is reasonable to ask whether the cultural Celticisms attributed to Norse Hebrideans, such as the family of Caittil Find/Ketill flatnefr (cf. Jennings and Kruse 2009b:126–133), should actually be read as aspirational if facile deference to the still powerful Gaelic-speaking rulers of Ireland proper—a kind of prestige by association (cf. Fellows Jensen 1996:120). Place-name continuity In the pre-map-making cultures of the early medieval Hebrides, place-names would have been created by local speech communities or “user groups” from the word material and onomastic grammar available at the point of creation. While we might expect a Gaelic speech community to have coined names in Gaelic, a Norse speech community would have created names using Old Norse material. Once those names had been coined, they would continue to exist within their respective user-group(s) as long as there was a need for them. When that need disappeared so too would the names (cf. Kruse 2004:102–103). As the survival of names, whether Gaelic or Norse, points to a certain amount of continuity in local user groups, it follows that the historical-philological study of them (cf. Sandnes 2003:109–111) could help to reveal developments in ethno-linguistic identity. In his groundbreaking survey of the place-names of Lewis, Captain W.F.L. Thomas (1874–1876:503) established that the ratio of Norse to Gaelic farmnames was 4:1. He observed, moreover, that the range and distribution of the Scandinavian material was such that it could only be explained in terms of the “extirpation”, or genocide,8 of the natives at the hands of the Norse during the Viking Age (Ibid.:503–504). The results of his subsequent assessment of Islay farm-names, on the other hand, were almost reversed, with the corresponding ratio being 1:2 (Thomas 1881–1882:273). Perhaps unsurprisingly, the discrepancy between these two ratios was quickly seized upon by adherents of the North versus South school as evidence of cultural continuity in the South. It should be stressed, however, that Thomas’ own conclusions on the Islay material were ambiguous and actually allowed for similarly unpleasant developments there too (Ibid.:273–276). A further area of concern, given the retrospective nature of Thomas’ work, is how little attention has been paid to the historical provenance of his source material. Both surveys were based on place-names recorded in late 19th-century County Valuation Rolls. With the best part of a millennium separating these particular toponymic snapshots from the height of the Viking Age, the scope for change in land-ownership, settlement distribution, and therefore place-names is enormous. At best, they should be treated as palimpsests. Following the demise of Norse culture in the Isles, most new material created is likely to have been in Gaelic or English and not Norse, resulting in a gradual but steady erosion of the proportion of Norse names (cf. Thomas 1874– 1876:504). It is worth noting here that just over a century earlier, on MacDougall’s 1749–1751 Map of the Island of Islay (Smith 1895:552–553), only around half of the linguistically certain farm-names are actually Gaelic in origin. Of those which are, around half can reasonably be considered post-Norse, either because they contain common habitative generics such as cill (G, chapel/graveyard) or baile (G, farm) A. Macniven 2013 Journal of the North Atlantic Special Volume 4 7 unlikely to pre-date ecclesiastic and administrative developments of the later Middle Ages (cf. Price 1963:119–124 and FN1), or because they are of the phrasal type considered by Watson (1904:xl) to be late. When we then consider the known waves of Gaelic-speaking immigration to Islay in the 12th, 16th, and 17th centuries, from Argyll, Northern Ireland, Nairnshire, and elsewhere (cf. Caldwell 2008, Storrie 1997), it is hard to imagine that the percentage of Norse material in the local nomenclature was not once substantially higher. When using the number and distribution of surviving Norse place-names to help assess the maximum extent of Norse settlement, it follows that the relative size of the corpus is probably less important than the typology and physical context of the individual names within it (cf. Thomas 1874–1876:503–504). The Norse habitative material on MacDougall’s map comprises a wide range of generics. Of the 31 certain Norse generics denoting settlement, however, only 11 are cultural: e.g., bólstaðir (18 examples), staðir (11), and land (4). The remaining 20 generics are topographic: e.g., dalr (11 examples), vík (6), and á (4). Between them, these 20 topographic generics account for around a quarter of all of the settlement-names shown on the map– pointing to an early phase of Scandinavian settlement where the landscape was described and apportioned from a Norse perspective without reference to pre-existing toponymic tradition (cf. Jennings and Kruse 2009a:87–93, Kruse 2005:141–144). By way of contrast, the majority of Gaelic settlement names shown on the map, amounting to around 30% of the total, have habitative generics, suggesting origins in a later, more mature phase of settlement when landholdings were sub-divided and/or re-named for their function or tenant rather than location. By plotting known settlement centers on a map and evaluating the surrounding landscape in terms of basic agricultural favorability, it is also clear that those with Norse names are fairly evenly spread across the whole island (Fig. 2), and no more likely to be associated with poor-quality land than their Gaelic-named counterparts. Contrary to observations by Olson (1983:134–176) and MacEacharna (1976:82–83), this kind of checkerboard pattern is less likely to point to early cultural hybridization than the existence of an island-wide Norse speech community. Application of Olsen’s (1934) User Group theory to the Islay place-name material leads to much the same conclusion. According to Olsen, all place-names can be assigned to one of three broad categories: gårdens navn, bygdens navn, and veiens navn (Nor: names of the farm, names of the district, and travellers’ names), each with its own range of user groups who create and maintain them. Thus, while the names of minor features on a given farm at a given point in time—such as streams, hollows, sheep-folds, etc.—might only be known to individuals living and working on that farm, those of more conspicuous features—such as larger hills, rivers, roads and the farms themselves—might be known to everyone in the district (Olsen 1934:10–12). “Names of the district” do not survive through isolation from the dominant speech community. They survive because they are used by it. Had the dominant speech community been Gaelic or mixed, we would expect this to be reflected in the form of the names. Instead, there are hints in the place-name record that the name-creating community was monolingual Norse. In Islay, as elsewhere in the Hebrides, there are a relatively large number of formally Gaelic names containing fossilized Old Norse naming elements, or ex nomine onomastic units (cf. Cox 1988–1989): e.g., Beinn Tart a’Mhill from Gaelic Beinn (hill) + backformed or grammatical Gaelic /t/ + Old Norse *Hartafjall (Stag Fell); Glen Egedale, from Gaelic Gleann (valley) + Old Norse *Eika(r)dalr (Oak Valley); and Sanagmore, from ON *Sandvík (Sandy Bay) + the Gaelic contrastive modified mór (large, greater). As the vocabulary and syntax of the Norse parts show no influence from Gaelic, they cannot be described as “hybrid names” in the sense that they were conscious products of a bi-lingual society (cf. Cox 1988–1989). They are, instead, more likely to represent names of an established Norse-speaking community which have been updated at some point after those communities came to speak Gaelic and the original meaning of the names had been forgotten. Conversely, and of crucial importance, there are no convincing Islay examples of dependent Norse names containing Gaelic ex nomine onomastic units, no *Glenmordalr or *Bailemartinsstaðir, suggesting that the cultural break at the beginning of the period of Norse settlement was rather more severe. This is not to suggest that the arrival of the Norse in Islay would have seen a complete disjuncture in the place-name record. The island-name Islay, which appears as “Ilea insula” in Adomnán of Iona’s late 7th-century Vita Columbae (Sharpe 1995:172) is a prime example of just such an early survivor. But given the low number of potentially pre-Norse names identified so far in Scotia Scandinavica (Gammeltoft 2007, Kruse 2005), their survival is probably best explained in terms of their usefulness as aides to navigation or bookkeeping by the incoming Norse, without necessarily implying peaceful or amicable interaction between the two population groups (cf. Kruse 2005:79–87). Similar conclusions can be drawn from the distribution of Norse nature-names. Although these currently comprise only a small fraction of the total Islay nomenclature, they are extremely overA. Macniven 2013 Journal of the North Atlantic Special Volume 4 8 ity of the most conspicuous features, including the bays, hills, and declivities tend to build upon Norse topographic generics. Port Alsaig, Frachdale, Maol Ghrasdail, Glen Astle, and Giol, for example, appear to derive from ON *Állsvík (bay), *Frakkadalr (valley), *Grasdalr (valley), *Ássdalr (ridge/valley) and *Gil (gully), respectively. represented when it comes to the more conspicuous topographical features. Take, for example, the west coast of the Oa peninsula. While there are numerous small features in the landscape here with transparently Gaelic names, there are very few which are conspicuous enough to be easily located without detailed description. By way of contrast, the major- Figure 2. Ultimate language background of non-English settlement-names on McDougall's map of 1749–1751 (after Macniven, in press). A. Macniven 2013 Journal of the North Atlantic Special Volume 4 9 further, barrel-shaped mound on Nave Island being indicative, perhaps, of the pre-requisite Norse longhouse (NMRS: NR27NE2). By way of contrast, the same overviews tend to massively underplay the relative proportion of smaller Scandinavian artifacts recovered from the area. Until the latter part of the 8th century, the language, cultural outlook, and projected identity of the Inner Hebrides were not only Gaelic but Christian (Fraser 2010, Woolf 2007). Following the arrival of the Vikings, however, there appear to have been radical changes to both of these defining features of local identity. Thus far, the remains of around forty pagan (i.e., furnished) Scandinavian graves have been found in the Hebrides as a whole (cf. Brown 1997, Graham-Campbell and Batey 1998, Holman 2009:127–128). The majority of these have come not from the Outer Hebrides, but the Inner Isles and adjacent mainland. In fact, for the more southerly islands such as Colonsay and Islay, these prestigious finds constitute the vast bulk of demonstrably Viking Age material. In Islay, they comprise the seven or more high-status pagan burials from Ballinaby (NMRS: NR26NW 4.0, 4.1, 4.2, 4.3 and 22), Newton Distillery (NMRS: NR36SW 2), and Cruach Mor (NMRS: NR35SW 1); to which we can add a silver hoard from Machrie Farm (NMRS: NR34NW 18); and a possible example of Scandinavian interlace sculpture from Dòid Mhàiri (NMRS: NR34NE 18). These finds dominate the archaeological record for this period and point to cultural disjuncture of a kind unlikely to have come about without the physical settlement of pagan Scandinavians (cf. Eldjárn 1984:7, Richards 1999). In fact, as the graves contain a relatively even mixture of men and women, it is tempting to see them as evidence for the settlement of Scandinavian communities. Far from representing an isolated and early development in the history of Norse-native interaction, it should also be noted that the majority of this material appears to date from the second half of the 10th century—around a century and a half after the first raid on Iona—suggesting the maintenance of revised cultural priorities over a period of at least several generations. From invasion to “predatory migration” Taken together, the linguistic, place-name, and archaeological evidence point to significant cultural change in Islay during the Viking Age. If it had been piecemeal but protracted or involved substantial hybridization, we would have expected the Norse material to have exhibited far greater interference from Gaelic. As things stand, the agency throughout appears to be Norse, and the outcome for the indigenous population devastating, a result that seems unlikely without the relatively large-scale It is important at this stage to ask how such breadth of settlement was achieved. Land, in the early medieval Gaidhealtachd, was a tightly controlled commodity. All land in Islay, whether under physical occupation or not, would have been legally owned and jealously guarded by a community unlikely to surrender it lightly to representatives of an aggressive ethnic “other”. The importance of landownership in maintaining the social order is stressed in a number of Early Irish law-tracts such as Críth Gablach (Kelly 1988, 1997). With the foundations of personal and political power lying, quite literally, in the land, the lack of land-ownership equated to a lack of status. Even when individuals had land to spare, giving it up or selling it could lead to a quantifiable loss of social standing (Gerriets 1983, 1987). Indeed, there seem to have been mechanisms in place to preserve vested interests in landed property. The text now known as the Senchus fer Alban, for example, includes a census and naval levy from 7th-century Dàl Riata (Bannerman 1974, Dumville 1997), suggesting that its rulers were well placed to fend off unwanted speculation. A Norse emigrant wanting to relocate to Islay during the Viking Age would have required the diplomatic standing to acquire land rights from local warlords through a process of gift-exchange, or the military strength to invade, settle, and subdue any subsequent reprisals at a local or regional level. As we shall see, below, the latter seems most probable. The lack of convincing material evidence for settlement It is regularly observed that Argyll and the Inner Isles have yet to produce a single demonstrably Scandinavian settlement (e.g., Barrett 2008b:415), the clear implication being that without the discovery of a typical Viking “longhouse”, there is little convincing evidence for large-scale immigration. For Islay, it would be equally pertinent to note that none of its Early Medieval monastic centers or dry-stone fortifications appear to have remained in use after the onset of the Viking Age (cf. Brown 1997:126) and that no settlement sites from this period have been identified from any architectural tradition, whether that be Scandinavian, Dalriadan, or otherwise. It might also be observed that this was also true until relatively recently of the Outer Hebrides, where sites like Bornais and Cille Pheadair are now producing Norse assemblages richer than most of those known from the Northern Isles (Sharples and Smith 2009). Norse antecedence has, moreover, been suggested for a number of ruins and mounds in Islay, including those at An Sithean (NMRS: NR26NE 3),9 Port Buidhe (NMRS: NR5NE 8), and Àird Thorrinnis (NMRS: NR26NW 5), with a A. Macniven 2013 Journal of the North Atlantic Special Volume 4 10 movement of native Scandinavian speakers into the area. For the past 60 years and more, however, this is a scenario that the archaeological community have been reluctant to countenance. Why this should be the case, given the general acceptance of Thomas’ late 19th-century invasion hypothesis for the Outer Isles, is not immediately obvious. The answer lies in world events of the early 20th century. On a surface level, there is little to separate Thomas’ thinking from the principles of “Culture- History” developed from the 1890s by theorists like Gustav Kossinna, which saw culture as a manifestation of biological heritage or “race” (e.g., Anthony 1990, Trafford 2000). As a result, it was something of a setback for Inner Hebridean Viking Studies when Kossinna’s views went on to capture the imagination of Nazi ideologists with well-known and horrific consequences. In the aftermath of the Second World War, the acceptability of concepts like invasion and even migration as explanatory models for cultural change went into serious decline. By 1966 and the publication of Grahame Clarke’s seminal article on “The Invasion Theory in British Archaeology”, the archaeological backlash to the Culture History paradigm was fully articulated. Rather than dwell on the unfathomable horror of man’s inhumanity to man, prehistorians preferred to look forward to a more optimistic past. Clarke’s focus was not on Scotland or the Viking expansion, but on southern England during the Neolithic and Bronze Age. His argument that cultural change should be seen as a largely internal process was, nevertheless, well received by colleagues involved in the study of Scotland’s Viking Age, and by the early 1970s, the discovery of Viking-style artifacts, houses, and even place-names north of the border was routinely explained in terms of the gradual or peaceful migration of ideas rather than the sudden and aggressive obliteration of one population group by another (cf. Bäcklund 2001, Owen 2004, Ritchie 1974; see also Barrett 2003, 2004). These piecemeal and pacifist approaches have not been without their detractors, most notably the Shetland archivist Brian Smith (2001) and the excavator of the Viking settlement at the Udal in North Uist, Iain Crawford (1981), but they have remained surprisingly resilient, despite contemporary examples of genocide and ethnic cleansing in Cambodia, Rwanda, the Balkans, and elsewhere (cf. Jones 2006:185–247). Now, after forty-five years of post-Clarke theorizing, the “Invasion” topos is once again a hot topic. This time, however, because it is being revisited and reworked by the historical community. Contextualized reappraisal of a wide range of known and suspected pre- and early historic migrations by Heather, Halsall, and others has led to a hybrid theory of migration and invasion which actually allows for the large-scale and aggressive movement of mixed population groups (e.g., Halsall 2007:417–454, Heather 2009:1–35). The circumstances in which this “predatory migration” is deemed possible still tend to limit the number and social mix of predatory migrants, but for economic rather than ideological reasons: the primary motivation being seen as access to wealth, albeit facilitated by the exploitation of political and economic structures rather than genocide. In his 2009 survey of “barbarian” migrations into the late Roman and early medieval West, Heather (2009:32) argues convincingly that the main limiting factor in the movement of people would have been the cost of transit. Thus, while the land-locked and large-scale economies of Late Antique Europe facilitated the slow movement of large and socially mixed groups of barbarians into the Roman Empire, the sea-borne, and therefore comparatively expensive nature of the Viking diaspora precluded largescale or low-level participation (Ibid.:452–504). At a conceptual level, this clearly favors the idea of “elite transfer” (Ibid.:23), whereby a limited number of high-status immigrants gain leading positions in their target communities. Indeed, independent reviews of the evidence from England and Normandy suggest that when migration does lead to cultural change, it is largely the result of “elite emulation” by the surviving locals, before the eventual assimilation of their new overlords (cf. Abrams 2013, Woolf 2007:292). According to Kastovsky (1992:324), just such a scenario can be inferred from the technical nature of Scandinavian loanwords in Old English (see also Townend 2005), borrowed as the English Danelaw took shape under a limited number of military architects. The volume and nature of loans into Middle English, on the other hand, suggests that subsequent immigration was not only large-scale but sustained, and led not to integration but social equilibrium with surviving parts of the native community (cf. Kastovsky 1992:325). Even so, it would be wrong to imagine that the flow of settlers was representative, democratic, or open-minded. As Heather (2009:486–487) points out, the bulk of first generation immigrants are likely to have been veterans of the heathen armies who acquired landed property by virtue of military prowess. Although it is quite possible that some of them would then have imported family and servants from Scandinavia, this would not always have made financial sense. Sea-going ships were an expensive commodity—expensive to build, expensive to crew, and expensive to maintain (cf. Bil 2008:170–171). It stands to reason, therefore, that passage on one of them would also have been expensive, and that freemen of meagre A. Macniven 2013 Journal of the North Atlantic Special Volume 4 11 In the late 8th century, however, perceptions of socio-economic value were rather different. On the eve of the Viking Age, Islay was economically and politically important enough to play host to the powerful cenèl nOengusa. Its contextually unusual expanses of fertile, limestone-derived soils (cf. Maltman et al. 2000, Wilkinson et al. 1907), and a strategic location between the treacherous waters of the Coire Bhreachan and the North Channel were no doubt as crucial to this early prominence as they were to its subsequent incarnation as the seat of the Gaelic-speaking Lords of the Isles. Considering the likely routes of transit between Norway and the Irish Sea during the intervening period, it seems improbable that these qualities would have gone unnoticed by wealth-seeking Scandinavians. Nevertheless, the available evidence does not point to the kind of elite take-over and subsequent cultural hybridization witnessed in Anglo-Saxon England. There are a number of potential reasons for this apparently different outcome. The issue of land-ownership has already been discussed. Although the situation here is unlikely to have been radically different from that encountered in England, problems of communication between speakers of Norse and Gaelic would no doubt have complicated matters. Unlike Old Norse and Old English, these two languages belong to different families—North Germanic and Celtic—and are unequivocally mutually unintelligible. Given the importance placed on ethnic association during this period, the inability to communicate with other groups could well have led to their demonization or even dehumanization, making early interaction not only difficult but undesirable. Had the Hebrides enjoyed a network of specialist centers, towns, and emporia like Ireland or England, appropriation of these resources would have facilitated interaction at arm’s length. With the settlement structure in Islay being exclusively rural, however, close contact with neighbors was unavoidable. In these circumstances, it would make sense for a military elite to clear away locally entrenched hostility and import more biddable farm-workers from elsewhere. It has been argued that relentless attacks by Norse Vikings in the two generations leading up to this point would have had a destabilizing effect on the Hebridean population, causstanding would have been unlikely to waste time and resources ferrying women, peasants, and slaves across the sea when they were already there to be had in large numbers. On the contrary, for the newly “landed”, socially ambitious, and wealth-hungry Danish veteran, marrying into an established local dynasty would have been a shrewd career move (cf. Hadley 2002:61), relatively unhindered by problems with intercommunication and greatly increasing the prospect of cultural hybridization. Perspectives on Islay With the physical distance between Norway and the Inner Hebrides being even greater, and the costs of transit between the two being presumably even higher, it is tempting to imagine an even more streamlined movement of people, with an even less disruptive impact on local population groups and traditions. Modern perceptions of the area appear to support this assumption. From a contemporary, mainland-based perspective, Islay is certainly peripheral,10 and with much of its 65,000 ha dominated by mountains and peat-bogs (Fig. 3), it can also be considered marginal— particularly when compared with other known targets for Norse settlement such as Man and Orkney. If the main criterion for Scandinavian migration was the acquisition of wealth, it would be difficult to see how an island like Islay might have piqued the interest of any Norse landnámsmen. Figure 3. Landscape types of Islay (adapted from SNH 1996:76). A. Macniven 2013 Journal of the North Atlantic Special Volume 4 12 ties like Islay. It is, moreover, possible that matters were compounded by regional collusion. With the focus of Scotland’s power-politics already shifting east, preoccupations are more likely to have been with the threat from expansionist Wessex than Viking raids. Norse settlement in the West could well have been tolerated or even encouraged as a convenient bulwark against aggressive Irish factions such as the cenèl nEoghain (cf. Woolf 2007:114–115). Indeed, native collusion in the creation of a Norse buffer zone may be reflected in saga accounts of Thorstein the red Ólafsson, who we are told in Laxdæla saga Chapter 4, “raided far and wide throughout Scotland and was everywhere victorious. Later he made a treaty with the Scots and became king over the half of Scotland they ceded to him” (Magnusson and Pálsson 1969:51). Although it seems extremely unlikely that either Thorstein, or a figure like him, ever controlled anywhere near half of Scotland as the author of Laxdæla saga might have known it, there is at least a faint possibility that this episode preserves the memory of a Norse partition of Dàl Riata leading to the perceived ethnic division between Innse Gall and Argyll (cf. Woolf 2004:94). There is no reason why the same fleets of ships involved in military expeditions and slaving might not also have filled the Isles with settlers. The suggestion that whole communities were relocated from Norway to further this kind of culturally imperialist agenda might seem far-fetched, but it is not without precedent in later times. We can point here to plantations in Ulster in the 16th century, Islay itself in the 17th, Palestine in the 20th, and many others. It might be argued that even this kind of development could have led to a relatively consensual process of Scandinavicization. As we have already seen, however, the evidence of material and linguistic artifacts alike speaks against this. In the face of a culturally dogmatic Norse invasion, the fate of any surviving locals is unlikely to have been pleasant. Dispossessed and subjugated with nowhere to run or hide, many may have found the situation unbearable. For nobles without status or land and apparently abandoned by regional overlords, perhaps the only honorable option was death by the sword—if not immediately, then through military service with the Norse in Ireland? The very short period over which the Gall Ghàidheil are said to have been active suggests they may have been a finite resource, the sons of the native Hebridean aristocracy sent overseas to die? Discussion Large-scale, mixed migration from Norway to the Inner Isles may have been logistically possible. ing many to flee and making those who remained far more pliant (cf. Jennings and Kruse 1995:292–293, Kruse 2005:151–152). But if resistance to Norse settlement had been particularly fierce, the deliberate disposal of fighting men would have been necessary to ensure the safety of the settlers,11 and could well have been achieved in fairly short order. The downfall of native society in this scenario would have been its island-based environment. Islay may be the second largest of the Inner Isles, but even so, it can hardly be considered big: no part of it is more than two hours’ walk from the sea. Had the Vikings wished it cleared, several strategically placed warships could have done the job in a day. There are indications that the resistance had been overcome by the middle of the 9th century. In the Annals of St Bertin for 847, for example, we learn of the “Northmen” getting control of all the islands around Ireland—in other words the Inner Hebrides—without encountering any resistance from anyone (Nelson 1991:65, Woolf 2007:100). This development may be reflected in later saga references to the Scandinavian “pirates” who were already resident in the Isles before the expedition of Ketil flatnose to subdue them (e.g., Pálsson and Ewards 1989:25–26). The likelihood of large-scale Scandinavian operations in the Isles around this time is also supported by near-contemporary references in the Irish annals to large Viking fleets in the Irish Sea: 140 ships in AU 849.6, 160 in AU 852.3 and 200 in AU 871.2—more than enough to clear the entire west coast. There has been some debate as to the accuracy of these accounts. While Sawyer (1971:17, 126) accepted reports of smaller fleets at face value, he regarded those numbering in the hundreds as hyperbole. As Smyth (1999:4–9) points out, however, there is no reason why equal credence should not be given to larger numbers, when both corroborative accounts and similar examples are found in the annals of continental Europe. Attention can also be drawn to the entry under AU 871.2, involving the removal of a “great prey” of slaves to Dublin by Amlaíb and Ímar following their siege of the fortress of Dumbarton Rock in Strathclyde. As similar endeavors are recorded in Ireland (cf. Holm 1986: 317–345, Hudson 1999:39– 66, Smyth 1999:21–22), it would be surprising if the same had not happened in the Hebrides. The scale of these slave raids is difficult to quantify. The figure of 3000 captives given in AU 951.3, for example, in conjunction with the raid on Kells in County Meath in Ireland, is no doubt exaggerated. It would have been familiar to scribes as a relatively common Biblical synonym for “a large number”. But the removal of even a fraction of this number would have had a devastating effect on small, island-based communiA. Macniven 2013 Journal of the North Atlantic Special Volume 4 13 It may also be reasonably consistent with the available evidence. Such a development would nevertheless be difficult to reconcile with the cost-benefit focus of current migration theory. If the main aim of emigration was to increase access to wealth, smallscale relocation would have been cost ineffective for ordinary Norwegians and therefore undesirable. But neither, when it came to the landed classes, would be paying to move large numbers of male farm workers and women, when these were already available in abundance in the target settlement zone. Surely the most effective way for mid-ranking nobles to benefit from foreign economies would be to integrate into, manipulate, and exploit them: whether this was by dominating local communities, as in England, or filling a niche in the market, as in Ireland? Two issues should be re-addressed here. The first of these is agency. Although it is popularly imagined that the majority of participants in the Viking expansion would have moved voluntarily and on their own initiative, it seems more likely in times when both social and geographical movement were tightly controlled that real agency in the migration process would have lain with the nobility. If we then accept that it was not access to wealth per se that was important but access to the social status this gave, the situation becomes immediately more tenable. Access to portable wealth appears to have played an important role in maintaining prestige status in the early Viking Age (Barrett 2008a:680– 681). Later on, however, as centralized authority became an entrenched part of the Norwegian political system, access to silver alone would no longer guarantee position. In circumstances like this, the prospect of losing status may have outweighed the cost of relocating the community that still respected it. When the middle men began to feel the pinch in 9th-century Norway, they did not travel to the Scottish Isles with a view to integrating into the native communities, they did so because they wanted to preserve their privileges. The best way to do that would be to recreate the societal ideal they had left behind. This kind of cultural imperialism is suggested by the typically Scandinavian long-houses discovered across the Northern and Western Isles and the pagan cremation burials identified at several sites from Arran to Orkney, but virtually nowhere else in the British Isles. Just such a scenario also finds striking parallels in a more recent wave of Norwegian migration. From the early 19th century until the early 20th, around 800,000 Norwegians left their native shores for a new life in North America. Although it might be assumed that their primary motivation for so doing was to exploit the financial opportunities offered by the American economy, the reality was not as straightforward. Judging from the insular, inwardlooking, and ethnically exclusive communities set up by Norwegian migrants to the mid-west of the USA (Munch 1949:780–781), it seems that a major reason for leaving was to maintain individual status built on traditional values. Norway itself was changing, and without the re-creation of traditional Norwegian communities abroad, this would have been impossible. This perspective begs the question of which changes might have provoked their ancestors to do likewise? Identifying Formative Push and Pull Factors For the Inner Hebrides, as elsewhere, it would be counter-productive to attribute all the events of the Viking Expansion to a single deterministic cause (cf. Barrett 2008b:671). Indeed, when comparing the evidence from Islay with other parts of the expansion zone, it is apparent that Scandinavian activity abroad, then as now, was situational and changed over time and place. Crawford (1987:39–48) has suggested, not unreasonably, that the changing nature of this activity could be seen as an unplanned reaction to changing circumstances in the Irish Sea region (pull factors). It would nevertheless be unwise to ignore contemporary currents in the social and political life of Scandinavia, some of which would have encouraged migrants to leave (push factors). The following three-stage model might help to explain the patterns of Norse-native interaction suggested by the evidence. Stage 1 Push. While the early raids would hardly have been possible without the coincidence of a number of interlinked phenomena, such as the introduction of the sail, these, in turn, are most likely to have been born out of, or employed as a direct result of social or economic crises. Of those postulated to date, perhaps the most compelling is the disruption of external revenue streams and resulting destabilization of the elite gift economy (e.g., Barrett 2008a, Hernæs 1997, Näsman 2000, cf. Samson 1991). For Norwegian warlords, hemmed in by the mountains of Kjølen to the east and the Danish political system to the south, the only opportunities for economic expansion lay in the west. Pull. At the same time, documented political instability in and around the Irish Sea (Fraser 2009, Woolf 2007) would have facilitated access to its accumulated wealth, proving attractive to disaffected Norwegian chieftains looking for new sources of income to improve their standing in socio-economic systems at home. The apparently seasonal nature of the early raids suggests that this was achievable without the need for a permanent foothold abroad. A. Macniven 2013 Journal of the North Atlantic Special Volume 4 14 Stage 2 Push. The success of stage 1 and a resultant, sudden influx of new wealth to Norway leads to a fatal economic imbalance. With more nobles now able to participate in the power stakes, the old order collapses into a violent spiral of political one-upmanship and centralization, conflated in the figure of Harald Finehair in the saga literature, and squeezing out the middle ranks. The appearance of named individuals in the Irish annals such as Saxolb (AU 837.9), Tuirgéis (AU 845.8), and Tomrair (AU 848.5), without any apparent links to central Norwegian dynasties, suggests that foreign resources are now more fiercely contested and need to be more directly controlled to be effectively exploited. Pull. The fractured Inner Hebridean seascape, destabilized over a generation or more of Scandinavian violence offers extreme logistical advantages for settlement compared to expansive and contiguous landscapes, such as Ireland, in establishing and maintaining the safety of Norse emigrant communities. But it is also close enough to a perpetually wartorn Ireland to give settlers scope to tap into further economic opportunities if required. The culturally imperialistic nature of this settlement sees the almost total subordination of the established Gaelic language and culture. Stage 3 A third and final stage of development can be postulated for the mature Scandinavian communities before changes in Norway lead to dwindling contact with the Isles and the demise of Norse culture in the Inner Hebrides. Push. The centralizing powers in Norway see the advantage of repeating this process in the “colonies”. From 853, powerful figures like Amlaíb (AU 853.2–FA 871) and Ímar (AU 857.1–AU 873.3) assert themselves over all the Scandinavians in Ireland and possibly also the Isles. Similarly, from the 850s and 860s, the Earls of Møre establish hegemony in Orkney. Once again, the middle-men are squeezed out, this time with the target of migration being the effectively virgin landscape of Iceland, and unquestionably involving large mixed population groups (Karlsson 2000). Pull. Ready-made constituencies of Norsespeakers— and later, land effectively free for the taking—draw and facilitate settlement. Concluding Remarks The idea of large-scale, mixed migration from Norway to the Inner Hebrides during the Viking Age finds resonance in the evidence from Islay. It also seems to be in broad agreement with what is known of the more general trends in the culture and politics of Early Medieval Scotland and Norway. Problems explaining such a development in terms of land-locked models of migration are countered to a certain extent by reference to the very different topographical and economic bases of the area in question. If we accept, moreover, that the main motivating factor in this particular movement was the maintenance of culture-specific social standing by disaffected mid-ranking chieftains, both the migration of mixed population groups to the Hebrides and the disappearance or extreme social subordination of the natives becomes even more convincing. That is not to say that pockets of Gaelic culture did not survive the Norse onslaught in the Inner Isles: the monastery of Columba on Iona being one very obvious example. Whether this was as anything other than specialist outposts, the result of local negotiation, or the interests of more Hibernofriendly Scandinavians in Ireland is difficult to say. In conclusion, however, one thing seems relatively certain: Islay before the Viking Age was a strategically important and economically powerful seat of Dalriadan chieftains near the geographical center of the Gaelic-speaking world. If cultural and linguistic disjuncture were possible here, there must have been exceptional reasons for any other part of the region escaping the same fate. Literature Cited Abrams, L. 2013. Early Normandy. Anglo-Norman Studies 35:45–66. Aðalbjarnarson, B. (Ed.). 1951. Snorri Sturluson: Heimskringla. Íslenzk Fornrit XXVIII, III. Hið Íslenzka Fornritafélag, Reykjavík, Iceland. Anthony, D. 1990. Migration in archaeology: The baby and the bathwater. American Anthropologist 92:895–914. Bäcklund, J. 2001. War or peace? The relations between the Picts and the Norse in Orkney. Northern Studies 36:33–47. Bannerman, J.W.M. 1974. Studies in the History of Dalriada. Scottish Academic Press, Edinburgh, UK. Barrett, J.H. 2003. Culture contact in Viking Age Scotland. Pp. 73–111, In J.H. Barrett (Ed.). Contact, Continuity, and Collapse: The Norse Colonization of the North Atlantic. Brepols, Turnhout, Belgium. Barrett, J.H. 2004. Beyond war or peace: The study of culture contact in Viking-Age Scotland. Pp. 207–218, In J. Hines, A. Lane, and M Redknap (Eds.). Land, Sea, and Home: Proceedings of a Conference on Viking- Period Settlement. Maney, Leeds, UK. Barrett, J.H. 2008a. What caused the Viking Age? Antiquity 82:671–75. Barrett, J.H. 2008b. The Norse in Scotland. Pp. 411–27, In S. Brink and N. Price (Eds.). The Viking World. Routledge, London, UK. Bartlett, R. 2001. Medieval and modern concepts of race and ethnicity. Journal of Medieval and Early Modern Studies 31(1):39–56. A. Macniven 2013 Journal of the North Atlantic Special Volume 4 15 Fellows Jensen, G. 1996. Language contact in Iceland: The evidence of the names. Pp. 115–124, In P.S. Ureland and I. Clarkson (Eds.). Language Contact across the North Atlantic: Proceedings of the Working Groups Held at University College, Galway (Ireland), August 29–September 3, 1992 and the University of Göteborg (Sweden), August 16–21, 1993. M. Niemeyer, Tübingen, Germany. Fellows Jensen, G. 2005. Extermination or Economic Exploitation? Pp. 100–118, In P. Gammeltoft, C. Hough, and D. Waugh (Eds.). Cultural Contacts in the North Atlantic Region: The Evidence of Names. NORNA, The Scottish Place-Names Society and The Society for Name Studies in Britain and Ireland, Lerwick, UK. Fraser, J.E. 2010. Early medieval Europe, the case of Britain and Ireland. Pp. 259–279, In D. Bloxham and D.A. Moses (Eds.). The Oxford Handbook of Genocide Studies. Oxford University Press, Oxford, UK. Friðriksson, A., and O. Vésteinsson. 2003. Creating a past: A Historiography of the Settlement of Iceland. Pp. 139–61, In J.H. Barrett (Ed.). Contact, Continuity and Collapse: The Norse Colonization of the North Atlantic. Brepols, Turnhout, Belgium. Gammeltoft, P. 2004. Scandinavian-Gaelic contacts. Can place-names and place-name elements be used as a source for contact-linguistic research? Nowele Vol. 44, March:51–90. Gammeltoft, P. 2007. Scandinavian naming-systems in the Hebrides: A way of understanding how Scandinavians were in contact with Gaels and Picts? Pp. 479–495, In B. Ballin Smith, S. Taylor, and G. Williams (Eds.). West Over the Sea: Studies in Scandinavian Sea-Borne Expansion and Settlement Before 1300. Brill, Leiden, The Netherlands. Gerriets, M. 1983. Economy and society: Clientship according to the Irish laws. Cambridge Medieval Celtic Studies 6:42–61. Gerriets, M. 1987. Kingship and exchange in pre-Viking Ireland. Cambridge Medieval Celtic Studies 13:39–72. Graham-Campbell, J. 1995. The Viking-Age Gold and Silver of Scotland (AD 850–1100). Edinburgh University Press, Edinburgh, UK. Graham-Campbell, J., and C.E. Batey. 1998. Vikings in Scotland. Edinburgh University Press, Edinburgh, UK. Guðmundsson, H. 1997. Um haf innan: Vestrænir menn og íslenzk menning á miðöldum. Háskálaútgáfan, Reykjavík, Iceland. Hadley, D. 2002. Viking and native: Re-thinking identity in the Danelaw. Early Medieval Europe II:I:45–70. Halsall, G. 2003. Warfare and Society in the Barbarian West, 450–900. Routledge, London, UK. Halsall, G. 2007. Barbarian Migrations and the Roman West 376–568. Cambridge University Press, Cambridge, UK. Heather, P. 2009. Empires and Barbarians. Pan Books, London, UK. Henderson, G. 1910. The Norse Influence on Celtic Scotland. J. Maclehose and Sons, Glasgow, UK. Hernæs, P. 1997. Storpolitikk og vikingtog på slutten av 700-tallet. Pp. 57–67, In I. Fuglesvedt and B. Myhre (Eds.). Konflikt i forhistorien (Aros Varia 30). Arkeologisk Museum i Stavanger, Stavanger, Norway. Holm, P. 1986. The slave trade of Dublin, ninth to twelfth centuries. Peritia 5:317–45. Bil, J. 2008. Viking ships and the sea. Pp. 170–180, In S. Brink and N. Price (Eds.). The Viking World. Routledge, London, UK. Borgström, C.-H. 1974. On the influence of Norse on Scottish Gaelic: Preaspiration of stops and pitch patterns. Norsk Tidskrift for Sprogvidenskap Supplementary Volume XI:91–103. Broderick, G., and Stowell, B. (Eds.). 1973. Chronicle of the Kings of Man and the Isles: Part I. George Broderick, Edinburgh, UK. Brøgger, A.W. 1929. Ancient Emigrants: A History of the Norse Settlements of Scotland. Clarendon Press, Oxford, UK. Brown, M. 1997. The Norse in Argyll. Pp 205–35, In G. Graham-Campbell (Ed.). The Archaeology of Argyll. Edinburgh University Press, Edinburgh, UK. Caldwell, D. 2008. Islay: The Land of the Lordship. Birlinn, Edinburgh, UK. Charles-Edwards, T.M. 1996. Irish warfare before 1100. Pp 26–50, In T. Bartlett and K. Jeffrey (Eds.). A Military History of Ireland. Cambridge University Press, Cambridge, UK. Clancy, T.O. 2008. The Gall-Ghàidheil and Galloway Journal of Scottish Name Studies 2: 19–50. Clarke, G. 1966. The invasion hypothesis in British archaeology. Antiquity XL:172–189. Cox, R.A.V. 1988–1989. Questioning the value and validity of the term “hybrid” in Hebridean place-name study. Nomina 12:1–9. Cox, R.A.V. 2002. The Gaelic Place-Names of Carloway, Lewis: Their Structure and Significance. Dublin Insti - tute for Advanced Studies, Dublin, Ireland. Crawford, B.E. 1987. Scandinavian Scotland. Leicester University Press, Leicester, UK. Crawford, I.A. 1981. War or peace:Viking colonisation in the Northern and Western Isles of Scotland. Pp. 259–269, In H. Bekker-Neilsen, P. Foote, et al. (Eds.). Proceedings of the Eighth International Viking Congress. Aarhus, Denmark. Downham, C. 2000. An imaginary Viking raid on Skye in 795? Scottish Gaelic Studies XX:192–196. Dumville, D.N. 1997. The Churches of Northern Britain in the First Viking Age. Fifth Whithorn Lecture, 14th September 1996. Friends of the Whithorn Trust, Whithorn, UK. Edwards, P., and H. Pálsson. 1989. Laxdæla Saga. Penguin, London, UK. Eldjárn, K. 1984. Graves and grave goods: Survey and evaluation. Pp. 2–11, In A. Fenton and H. Pálsson (Eds.). The Northern and Western Isles in the Viking World: Survival, Continuity, and Change. John Donald, Edinburgh, UK. Eliasson, S. 2000. Typologiska och areallingvistiska aspekter på de nordeuropeiska språkens fonologi. Pp. 21–71, In E.H. Jahr. (Ed.). Språkkontakt: Innverknaden frå nedertysk på andra nordeuropeiska språk. Nordisk Ministerråd, Copenhagen, Denmark. Etchingham, C. 1996. Viking Raids on Irish Church Settlement in the Ninth Century: A Reconsideration of the Annals. Department of Old and Middle Irish, St Patrick's College, Maynooth, Ireland. A. Macniven 2013 Journal of the North Atlantic Special Volume 4 16 MacEacherna, D. 1976. The Lands of the Lordship: The Romance of Islay’s Names. Argyll Reproductions, Port Charlotte, UK. Macniven, A. In press. The Vikings in Islay: The Place of Names in Hebridean Settlement History. Birlinn, Edinburgh, UK Magnusson, M., and Pálsson, H. 1969. Eyrbyggja Saga. Penguin, London, UK. Maltman, A., R. Elliot, et al. 2000. Guide to the Geology of Islay. Institute of Geography and Earth Sciences, Aberystwyth, UK. Manning, C. 1995. Early Irish Monasteries. County House, Dublin, Ireland. Manning, C. 2000. References to church buildings in the Annals. Pp. 37–52, In A.P. Smyth (Ed.). Seanchas: Studies in Medieval Irish Archaeology, History and Literature in Honour of Francis J. Byrne. Four Courts Press, Dublin, Ireland. Marsden, J. 2000. Somerled and the Emergence of Gaelic Scotland. Tuckwell Press, East Linton, UK. Martin, M. 2002. A Description of the Isles of Scotland Circa 1695. Pp. 12–31, In W.J. Withers, and R.W. Munro (Eds.). A Description of the Isles of Scotland Circa 1695, A Voyage to St Kilda: Martin Martin; with A Description of the Occidental, i.e., Western Isles of Scotland: Donald Monro. Birlinn, Edinburgh, UK. McDonald, R.A. 1997. The Kingdom of the Isles: Scotland’s Western Seaboard, c.1100–c.1336. Tuckwell Press, East Linton, UK. Munch, P.A. 1949. Social adjustment among Wisconsin Norwegians. American Sociological Review 14(6):780–787. Myhre, B. 2000. The early Viking Age in Norway. Acta Archaeologica 71:35–47. Näsman, U. 2000. Raids, migrations, and kingdoms: The Danish case. Acta Archaeologica 71:1–7. Nelson, J. 1991. The Annals of St-Bertin. Manchester University Press, Manchester, UK. Nieke, M. 1983. Settlement patterns in the 1st millennium AD: A case study of the island of Islay. Pp. 95–115, In J.C. Chapman, and H.C. Mytum (Eds.). Settlement in North Britain 1000 BC–AD 1000 : Papers Presented to George Jobey, Newcastle upon Tyne, December 1982. B.A.R., Oxford, UK. Nieke, M. 1984. Settlement patterns in the Atlantic Province of Scotland in the 1st millennium AD: A study of Argyll. Unpublished Ph.D. Dissertation. Glasgow University, Glasgow, UK. Nieke, M.R. 1990. Fortifications in Argyll: Retrospect and future prospect. Pp. 131–142, In I. Armit (Ed.). Beyond the Brochs. Edinburgh University Press, Edinburgh, UK. Nieke, M.R., and Duncan, H.B. 1988. Dalriada: The establishment and maintenance of an Early Historic Kingdom in northern Britain. Pp. 6–21, In S.T. Driscoll and M.R. Nieke (Eds.). Power and Politics in Early Medieval Britain and Ireland. Edinburgh University Press, Edinburgh, UK. Ó Corráin, D. 1998a. The Vikings in Ireland and Scotland in the ninth century. Pp. 296–339, In H.B. Clarke, M. Ní Mhaonaigh, and R. Ó Floinn (Eds.). Ireland and Scandinavia in the Early Viking Age. Dublin, Ireland. Holman, K. 2009. The A to Z of the Vikings. Scarecrow Press, Plymouth, UK. Hudson, B.T. 1999. The changing economy of the Irish Sea province: AD 900–1300. Pp. 39–66, In B. Smith (Ed.). Britain and Ireland, 900–1300: Insular Responses to Medieval European Change. Cambridge University Press, Cambridge, UK. Jackson, K. 1951. “Common Gaelic”: The evolution of the Goidelic languages. Proceedings of the British Academy 37:71–97. Jennings, A. 1996. Historical and linguistic evidence for Gall-Gaidheil and Norse in Western Scotland. Pp. 61–73, In P.S. Ureland, and I. Clarkson (Eds.). Language contact across the North Atlantic : proceedings of the working groups held at University College, Galway (Ireland), August 29–September 3, 1992 and the University of Göteborg (Sweden), August 16–21, 1993. M. Niemeyer, Tübingen, Germany. Jennings, A. 1998. Iona and the Vikings: Survival and continuity. Northern Studies 33:37–54. Jennings, A., and A. Kruse. 2005. An ethnic enigma: Norse, Pict, and Gael. Pp. 251–63, In A. Mortensen and S.V. Arge (Eds.). Viking and Norse North Atlantic, Selected Papers from the Proceedings of the Fourteenth Viking Congress, 19–30 July 2001. Tórshavn, The Faroe Islands. Jennings, A., and A. Kruse. 2009a. One coast—three peoples: Names and ethnicity in the Scottish West during the early Viking period. Pp. 75–102, In A. Woolf (Ed.). Scandinavian Scotland: Twenty Years After. The Proceedings of a Day Conference held on 19 February 2007, The Committee for Dark Age Studies, The University of St. Andrews, St. Andrews, UK. Jennings, A., and Kruse, A. 2009b. From Dàl Riata to the Gall Ghàidheil. Viking and Medieval Scandinavia 5:123–150. Johnston, A.R. 1991. Norse settlement in the Inner Hebrides ca. 800–1300 with special reference to the islands of Mull, Coll, and Tiree. Unpublished Ph.D. Dissertation. University of St. Andrews, St. Andrews, UK. Jones, A. 2006. Genocide: A Comprehensive Introduction. Routledge, London, UK. Karlsson, G. 2000. Iceland’s 1100 Years: The History of a Marginal Society. Mál og menning, Iceland. Kastovsky, D. 1992. Semantics and vocabulary. Pp. 290– 407. In R.M. Hogg (Ed.). The Cambridge History of the English Language. Vol. 1. The Beginnings to 1066. Cambridge University Press, Cambridge, UK. Kelly, F. 1988. A Guide to Early Irish Law. Dublin Institute for Advanced Studies, Dublin, Ireland. Kelly, F. 1997. Early Irish Farming: A study based mainly on the law-texts of the 7th and 8th centuries AD. Dublin Institute for Advanced Studies, Dublin, Ireland. Kruse, A. 2004. Norse topographical settlement names on the western littoral of Scotland. Pp. 99–109, In J. Adams, and K. Holman. (Eds.). Scandinavia and Europe 800–1350: Contact, Conflict, and Coexistence. Brepols, Turnhout, The Netherlands. Kruse, A. 2005. Explorers, raiders, and settlers. The Norse impact upon Hebridean place-names. Pp. 141–156, In P. Gammeltoft, C. Hough, and D. Waugh (Eds.). Cultural Contacts in the North Atlantic Region: The Evidence of the Names. Norna, Scottish Place-Name Society and Society for Name Studies in Britain and Ireland, Lerwick, UK. A. Macniven 2013 Journal of the North Atlantic Special Volume 4 17 Sellar, W.D.H. 1966. The origins and ancestry of Somerled. The Scottish Historical Review 45:123–142. Sharpe, R. (Ed.). 1995. Adomnán of Iona: Life of St Columba. Penguin Classics, London, UK. Sharpe, R. 2000. The thriving of Dalriada. Pp. 47–61, In S. Taylor (Ed.). Kings, Clerics, and Chronicles in Scotland 500–1297. Four Courts Press, Dublin, Ireland. Sharples, N., and R. Smith. 2009. Norse settlement in the Western Isles. Pp. 103–130, In A. Woolf, (Ed.). Scandinavian Scotland—Twenty Years After. The Proceedings of a Day Conference held on 19 February 2007, The Committee for Dark Age Studies, The University of St. Andrews, St. Andrews, UK. Shetelig, H. 1954. The Viking graves. Pp. 67–111, In A.O. Curle, M. Olsen, and H. Shetelig (Eds.). Civilisation of the Viking Settlers in Relation to their Old and New Countires: Viking Antiquities in Great Britain and Ireland Part 6. Aschehoug, Oslo, Norway. Sigurdsson, G. 1988. Gaelic influence in Iceland. Studia Islandica 46. Bókarútgáfa Menningarsjóðs, Reykjavík, Iceland. Smith, B. 2001. The Picts and the martyrs or did Vikings kill the native population of Orkney and Shetland? Northern Studies 36:7–32. Smith, G.G. (Ed.). 1895. The Book of Islay: Documents Illustrating the History of the Island. Private Publication, Edinburgh, UK. Smyth, A. 1977. Scandinavian Kings of the British Isles, 850–880. Oxford University Press, Oxford, UK. Smyth, A.P. 1984. Warlords and Holy Men. Edinburgh University Press, Edinburgh, UK. Smyth, A.P. 1999. The effect of Scandinavian raiders on the English and Irish churches: A preliminary reassessment. Pp. 1–38, In B. Smith (Ed.). Britain and Ireland, 900–1300: Insular Responses to Medieval European Change. Cambridge University Press, Cambridge, UK. Stahl, A.-B. 1999. Place-names of Barra in the Outer Hebrides. Unpublished Ph.D. Dissertation. The University of Edinburgh, Edinburgh, UK. Stewart, T.W. 2004. Lexical imposition. Old Norse vocabulary in Scottish Gaelic. Diachronica 21.2: 393–420. Storrie, M. 1997. Islay: Biography of an Island. The Oa Press, Isle of Islay, UK. Stuart, J., et al. (Eds.). 1878–1908. The Exchequer Rolls of Scotland. Edinburgh, UK. Sveaas Andersen, P. 1991. Norse settlement in the Hebrides: What happened to the Natives and what happened to the Norse immigrants? Pp. 131–47. In I. Wood and N. Lund (Eds.). People and Places in Northern Europe 500–1600.The Boydell Press, London, UK. Thomas, W.F.L. 1874–1876. Did the Norsemen extirpate the inhabitants of the Hebrides in the ninth century? Proceedings of the Society of Antiquaries of Scotland 11:472–507. Thomas, W.F.L. 1881–1882. On Islay place-names. Proceedings of the Society of Antiquaries of Scotland 16:241–76. Thomason, S.G. 1987. On mechanisms of interference. Pp. 181–207, In S. Eliasson and E.H. Jahr (Eds.). Language and its Ecology: Essays in Memory of Einar Haugen. Mouton de Gruyter, Berlin, Germany and New York, NY, USA. Totten, S., and P.R. Bartrop (Eds.). 2009. The Genocide Studies Reader. Routledge, London, UK. Ó Corráin, D. 1998b. Viking Ireland—afterthoughts. Pp. 421–452, In H.B. Clarke, M. Ní Mhaonaigh, and R. Ó Floinn (Eds.). Ireland and Scandinavia in the Early Viking Age. Dublin, Ireland. Ó Corráin, D. 2008. The Vikings and Ireland. Pp. 428–433, In S. Brink and N. Price (Eds.). The Viking World. Routledge, London, UK. Oftedal, M. 1954. The village names of Lewis in the Outer Hebrides. Norsk tidskrift for sprogvidenskap XVII:363–409. Oftedal, M. 1961. On the frequency of Norse loanwords in Scottish Gaelic. Scottish Gaelic Studies, Vol. IX, Part I:116–135. Olsen, M. 1926. Ættegård og helligdom. Norske stedsnavn sosialt og religionshistorisk belyst. Aschehoug and Co., Oslo, Norway. Olsen, M. 1934. Hvad våre stedsnavn lærer oss. Universitetets radioforedrag, Oslo, Norway. Olson, D. 1983. Norse Settlement in the Hebrides. Hovedoppgave Thesis. University of Oslo, Oslo, Norway. Owen, O. 2004. The Scar boat burial and the missing decades of the early Viking Age in Orkney and Shetland. Pp. 3–33, In J. Adams and K. Holman (Eds.). Scandinavia and Europe 800–1350: Contact, Conflict, and Coexistence. Brepols, Turnhout, The Netherlands. Pálsson, H., and P. Edwards (Trans., Eds.). 1972. The Book of Settlements: Landnámabók. University of Manitoba Press, Winnipeg, Canada. Pálsson, H., and P. Edwards (Trans., Eds.). 1989. Eyrbyggja Saga. Penguin, London, UK. Power, R. 1990. Scotland in the Norse sagas. Pp. 13–24, In G. Simpson (Ed.). Scotland and Scandinavia. John Donald, Edinburgh, UK. Price, L. 1963. A note on the use of the word baile in place-names. Celtica VI:119–126. Renfrew, C., and P. Bahn. 2000. Archaeology: Theories, Methods, and Practice. Thames and Hudson, London, UK. Richards, J.D. 1999. The Scandinavian settlement. Pp. 194–209, In J. Hunter and I. Ralston (Eds.). The Archaeology of Britain. Routledge, London, UK. Ritchie, A. 1974. Pict and Norseman in northern Scotland. Scottish Archaeological Forum VI:23–36. Samson, R. (Ed.). 1991. Social Approaches to Viking Age Studies. Tuckwell Press, Glasgow, UK. Sandnes, B. 2003. Dannelse og utvikiling av norrønne stedsnavne på Orknøyene. Det historiske-filosofiske fakultet, Norges teknisk-naturvetenskapelige universitet, Trondheim, Norway. Sawyer, P.H. 1971. The Age of the Vikings. E. Arnold, London, UK. Scales, L. 2010. Central and late medieval Europe. Pp. 280–303, In D. Bloxham and D.A. Moses (Eds.). The Oxford Handbook of Genocide Studies. Oxford University Press, Oxford, UK. Schulze-Thulin, B. 2001. Notes on the Old and Middle Irish loanwords in Old Norse. Nowele 39(September): 53–84. Scottish Natural Heritage. 1996. Landscape assessment of Argyll and the Firth of Clyde. Scottish Natural Heritage Review No. 78. Available online at http:// www.snh.org.uk/publications/on-line/LCA/argyll.pdf. Accessed 21 December 2011. A. Macniven 2013 Journal of the North Atlantic Special Volume 4 18 Townend, M. 2005. Language and History in Viking Age England: Linguistic Relations between Speakers of Old Norse and Old English. Brepols, Turnhout, The Netherlands. Trafford, S. 2000. Ethnicity, migration theory, and the historiography of the Scandinavian settlement of England. Pp. 17–42, In D.M. Hadley and J.D. Richards (Eds.). Cultures in Contact: Scandinavian Settlement in England in the Ninth and Tenth Centuries. Brepols, Turnhout, The Netherlands. Wamers, E. 1993. Insular finds in Viking Age Scandinavia and the state formation of Norway. Pp. 37–42, In H.B. Clarke, et al. (Eds.). Ireland and Scandinavia in the Early Viking Age. Four Courts Press, Dublin, Ireland. Ward-Perkins, B. 2000. Why did the Anglo-Saxons not become more British? The English Historical Review 115(462):513–533. Watson, W.J. 1904. Place Names of Ross and Cromarty. Northern Counties Printing and Publishing, Inverness, UK. Watson, W.J. 1926 (reprinted 1993). The Celtic Placenames of Scotland. Birlinn, Edinburgh, UK. Weinreich, U. 1953 (reprinted 1974). Languages in Contact: Findings and Problems. Mouton, The Hague, The Netherlands. Wilkinson, S.B., J.J.H. Teal, et al. 1907. Memoirs of the Geological Survey: Scotland: The geology of Islay including Oronsay and portions of Colonsay and Jura. Explanations of Sheets 19 and 27, with the western part of Sheet 20. HMSO, Glasgow, UK. Woolf, A. 2004. The age of sea-kings: 900–1300. Pp. 94–99, In D. Omand (Ed.). The Argyll Book. Birlinn, Edinburgh, UK. Woolf, A. 2007. From Pictland to Alba 789–1070. The University of Edinburgh Press, Edinburgh, UK. Endnotes 1The term “Viking Age” is a nebulous one, whose temporal span and implications vary from region to region. Here, it will cover the period of pagan Scandinavian raiding, settlement and cultural influence from the time of the first recorded raids in the area in the last decade of the 9th century to the “official” Conversion of the neighboring Northern Isles by Óláfr Tyrggvason in the last decade of the 10th century. 2AU = The Annals of Ulster, AI = The Annals of Innisfallen. FA = The Fragmentary Annals. All references to Irish annals are to University College Cork’s online Corpus of Electronic Texts http://celt.ucc.ie/index.html (Accessed 30 September 2011). ASB = The Annals of St Bertin (Nelson 1991). 3The supposed raid on Skye in AU 795.3 has been convincingly dismissed by Clare Downham (2000:192–196) as a scribal error for an earlier scríne (shrines) adding further detail to the preceding account of the burning of the monastery of Rechru (Rathlin Island). 4While it would be wrong to imagine that the recorded “Viking” raids were not short-lived or violent, and did not involve the non-consensual taking of precious objects (cf. Ó Corráin 1998b:438–439, Wamers 1998:37–72), it seems likely from the better-documented Irish experience that there were other targets, both monastic and secular, and that more sophisticated, political agendas were also in play. The monastic raid, for example, was a tried and tested feature of internecine power-struggles in pre- Viking Age Ireland (Charles-Edwards 1996, Etchingham 1996:15, Manning 2000:47–49, Ó Corráin 1998a:430). In addition to asserting dominance, burnings and other forms of destruction may have had more subtle functions. Fellows-Jensen’s (2005:109–133) survey of Scandinavian settlement in La Hague, Normandy, suggests that the incoming land-takers deliberately destroyed title deeds and tax lists kept by the Church to complicate subsequent legal challenges. 5Viking activity is not taken to include post-Conversion Scandinavian aggression such as Magnús Bareleg Ólafsson’s expedition of 1098, made famous by Norse skáld Björn krepphendi (ON, the Cripple-Handed) in Snorri Sturluson’s Magnúss saga berfætts (Aðalbjarnarson 1951:219–223). 6Terrimotus in Ili .ii. id Aprilis (AU 740.3), Death of Manx king Godred Crovan Haraldsson “in insula quae vocatur Yle” (Chronicle of the Kings of Man and the Isles §23 = Broderick and Stowell 1973). 7For ease of reference, the term “(Old) Norse” will be used in this paper to denote the varieties of Common Scandinavian spoken by Norwegian communities during the Viking Age, “Scandinavian” will be used when it is necessary to expand the linguistic scope geographically– e.g., to the dialects spoken by ethnic Danes during the same period. The term “Gaelic”, will be used primarily to denote the varieties of Goidelic dialect spoken in the Scots (= Irish) kingdoms of Dàl Riata and Ireland, whether before this point or more recently. 8For definitions of “genocide” and related terms such as “ethnic cleansing” see chapters 1–3 of Totten and Bartrop (2009). Further medieval examples are discussed in Fraser (2010) and Scales (2010). 9All entries in the National Monuments Record of Scotland (NMRS) can be accessed via CANMORE, the online database of The Royal Commission on the Ancient Historical Monuments of Scotland (RCAHMS): http://www. rcahms.gov.uk/canmore.html (Accessed 30 September 2011). 10In earlier times, Islay was erroneously believed to be the furthest west of all of the islands of Britain. On his visit to the island in the 1690s, the traveller Martin Martin was told by locals that the “village” of Coull on the west coast was so called by ancients because it represented the “back [part] of the world” (Martin 2002:275–276). 11Cf. Ward-Perkins’ (2000) discussion of the fate of the native Britons in England in the face of an earlier wave of settlement by Angles and Saxons.