Journal of the North Atlantic M. Thacker 2015 Special Volume 9 45 Introduction Although lime mortar first appears in the archaeological record of southern Britain during the Roman Iron Age (Williams 2004:4), it is the evidence for building with stone during that time which has often been emphasized (e.g., Parsons 1990, Pearson 2006). This emphasis has become so thoroughly inscribed into many insular archaeological interpretations that any subsequent association of people with stone buildings is now seen to have an implicit “Romanitas” (Ó Carragáin 2005:101). This concept has proven very useful for understanding medieval archaeological and historical material in suggesting that as Rome continued to be the center of the Church in western Europe, so “Catholicity” needed to be expressed and mediated by monumental masonry architecture (Gem 1983). Masonry, therefore, expresses identity. In England, this enabled Bede (731 [1990]) to assert nationality by comparing early medieval “Irish”, Pictish, and British timber ecclesiastical buildings with the contemporary stone-built churches of Roman Catholic Northumbria. In this early period of ecclesiastical reform, Bede and Wilfred were constructing a complex dialectic whereby stone means permanence, investment, strength, progress, civilization, and pan-continental, orthodox, universal Roman-Christian culture, in opposition to the particular, misguided, peripheral, and ultimately ephemeral cultural practices from which timber or turf structures had emerged. The Scottish “Atlantic Province” (or “Atlantic Scotland”), however, although relatively unaffected by the Iron Age Romanization of landscapes further south, has also been largely defined by its earlier and broadly contemporary archaeological evidence for sophisticated monumental stone buildings (Piggott 1966:7). This region was one of the four areas of Scotland classified by Piggott on the criteria of “geography, chronology, and culture” (Ibid.:3). In this scheme, the lands within Atlantic Scotland were interpreted as forming a coherent Iron Age province, evidenced by the extensive remains of stone-built structures found across the coastal north and west of the country—from the firth of Clyde to just north of Inverness, including the Northern and Western Isles (ibid., see also Lamb 2008). Lime mortar, however, does not generally appear in the archaeological record here until the early 2nd millennium A.D., coinciding not only with another period of widespread Church reform across much of Europe, but also with the Christianization of the Late Norse elite throughout the whole North Atlantic. This conversion was a great victory for the Church in extending its authority further beyond the former bounds of Imperial Rome (cf. O’ Reilly 2009), and this early lime evidence emerges when most domestic buildings here were indeed constructed of dry-stone, turf, and/or timber. Often, however, the substantial remains of Pictish, Iron Age, and earlier prehistoric stone structures would still have been the most visible and certainly the most monumental architectural forms within many settlement landscapes at this time (Driscoll 1998, Parker Pearson et al. 2004). In the archaeological record of this province, therefore, it is the appearance of lime-bonded or plastered masonry, rather than stone, which somehow architecturally defines this period of Christianization, as well as perhaps displaying something of the concepts of Romanitas and reform seen elsewhere (Ó Carragáin 2003:151). This is an interpretation that requires us to look closer at the lime itself and investigate its affect upon any pre-existing building traditions, and the wider built environment, during this and later periods. This province, however, is also defined by its geography, and although the archaeology suggests most medieval European, British, and even Scottish Cille Donnain Revisited: Negotiating with Lime Across Atlantic Scotland from the 12th Century Mark Thacker* Abstract - This paper introduces the initial findings from an ongoing buildings archaeology research project. I present the results of two wide-ranging mortar surveys which describe the archaeological evidence for lime mortars within the 12thcentury bicameral chapels of Atlantic Scotland and pre-Reformation chapels of the Western Isles. The diverse evidence for materials, plan form, and structure suggest important processes of cultural negotiation. I argue that lime mortar was an animated technological agent for Christianization, which enabled power and perspective to be mediated by complex culturally appropriate techniques. Special Volume 9:45–66 2010 Hebridean Archaeology Forum Journal of the North Atlantic *27 Upper Carloway, Isle of Lewis, Scotland, UK; m.thacker@ed.ac.uk. 2015 Journal of the North Atlantic 46 M. Thacker 2015 Special Volume 9 west-coast mainland building-limes were created by burning limestone or chalk with wood or coal in various kiln types (e.g., Cramp 2006, Hughes and Cuthbert 2000), Atlantic Scotland presents a range of regional environments, many of which might challenge this technological model. Limestone, woodland, and coal were available in some regions, but in others, including the Western Isles, archaeological evidence for 900 years of lime-mortar activity is evident within past landscapes without any known outcrop of limestone,1 little woodland, and no coal (Armit 1996:23, Withrington and Grant 1983:58). Moreover, focusing further on the Western Isles, very little archaeological evidence of lime-burning or mixing sites of any period have been recorded here. Later historical evidence documents the importation of lime, limestone, and coal from the late 18th century (Headrick [1800] 1870:17, Knox [1786] 1975:148), but also from this period to within living memory, building-limes were reported to have been produced on a small scale, by burning cockle shells in a peat-fired clamp—a procedure which would also leave little upstanding archaeology (MacIntyre 1993; Macqueen 1791–1799:145; Monro 1791–1799:42; Thacker, in press). Pre-modern lime provenances in the Western Isles, however, are not very well documented, although suggested possible carbonate sources have included shell sand (Reece 1981:15–16), shell (Young 1955), marl (Barrowman 2008:11) and mainland limestone (Macleod 1998:5). While this lack of earlier documentation and recorded physical evidence of production sites, particularly in the Late Norse and medieval periods, places greater reliance on the surviving material remains within the buildings themselves, most surveyors have been content to ambiguously apply the “shell-” prefix to Western Isles lime mortars, without further clarification of whether this refers to the binder, or to an aggregate material which also often contains a high shell content (Thacker 2012). A better archaeological understanding of individual lime mortars requires a more focused initial mortar-research agenda, and a more holistic approach to examining the different environments, buildings, mortars, and materials of the whole region, and ultimately the wider province. In this regard, the landscapes surrounding pre-Reformation chapels are particularly significant as places where material evidence of early and continuing lime activity can often still be observed. While the limited number of upstanding chapel remains and graveyard sites have posed considerable challenges to scholarship and dating (Crawford 1987:183), the ruinous condition of the buildings, mix of mortar material provenances, and surviving landscape contexts also make mortar archaeology an accessible and yet largely neglected tool with which to explore these places as well as the techniques from which they emerged. From the beginning of this project, Fleming and Woolf’s (1992) excavations and interpretation of the chapel at Cille Donnain in South Uist provided an important study in three respects: Firstly, building on the work of Cant (1975, 1984), this study extended the apparent regionality of broadly 12th-century bicameral (two-cell) chapels from the Northern Isles and Caithness to the Western Isles, and further suggested that this chapel planform was in these instances a particularly Norse regional expression. This association with ethnicity might appear curious given that bicamerality, whereby various masonry structures divide the nave from a relatively smaller chancel and so serve to more emphatically separate the congregation from the celebrant during mass2, is a common northern European articulation of Christian reform (see also Raven 2005:178), but this Norse interpretation brings some much needed complexity to our earlier medieval and Roman constructional dialectic, and will be further discussed below. Secondly, for the archaeologist in the field this use of masonry, wherever found, is very useful even where, as at Cille Donnain, the upstanding archaeology is particularly fragmentary. This site’s building footprint is distinctive and, although not an absolute rule, the majority of these chapels were built between the late 11th and early 13th centuries. The third important point to emerge from the Cille Donnain paper was an interpretation that this chapel’s relatively large size and landscape similarity to Finlaggan implied a high-status church. This conclusion is also curious, however, when the chapel was interpreted as dry-stone and in contrast with the recorded lime evidence found at many of the other bicameral chapels within this group as presented below. This then is the point from which this paper departs to introduce the archaeological potential and meanings of lime mortars across these regions. Our evidence lies within a province with various environmental conditions and material sources. Without at this stage discussing previous work that has convincingly demonstrated various Iron Age regionalities here (Armit 1991, Harding 2000, Hingley 1992), these landscapes do appear to have previously supported settlements with some degree of conformity in built culture: sophisticated monumental dry-stone masonry remains from earlier periods, contemporary Norse turf-building traditions, and Journal of the North Atlantic M. Thacker 2015 Special Volume 9 47 emerging masonry chapel forms. The intention is to explore how, where, and when different lime mortars are evidenced within the archaeological record from this point onwards: to assess the potential of more focused, non-intrusive, on-site mortar-archaeology methodologies at specific sites; compare inter-site mortar evidence to establish initial typological frameworks; and ultimately enable a more informed discussion of why this agent was, or was not, adopted by local communities at all. Lime mortar and bicamerality may have expressed and mediated Christianization, Romanitas, reform, and identity in the Late Norse period, but how were the architectural affordances enabled by these technologies negotiated within the emerging cultural landscape of this and later periods? Methodology and Results A multi-scalar framework was provided by two mortar surveys: Survey 1 investigated broadly 12th-century bicameral chapels across the former Scottish Atlantic Province to provide some chronological and plan-form uniformity for geographical–structural comparison. This survey was designed to investigate how the material fabric of chapels from different regions relates to plan-forms which appear to be reproduced across diverse landscapes. This survey began by undertaking a search of the 2687 “chapel” and 8232 “church” entries on the R.C.A.H.M.S. CANMORE website (www.rcahms.gov.uk). Further documentary sources such as local SMRs and archaeological papers were consulted where identified. Details of chapel plan-forms and any dating evidence were refined to include only those chapels of possible 12th-century construction in the local authority areas of Shetland, Orkney, Highland, Western Isles, and Argyll and Bute, and these buildings were further scrutinized for bonding type, coatings, wall thicknesses, phasing evidence, inclined jambs, and vaulting or arches. Survey 2 investigated only Western Isles chapels, but throughout the whole of the pre-Reformation period, with selected post-Reformation examples also, to provide some geographical uniformity for chronological–structural comparison. This survey began with the same sources as Survey 1, but the documentary information was then further resolved by field survey at the majority of any upstanding sites, in order to verify the evidence and provide a landscape context for the buildings and a masonry context for the mortars. As I have described elsewhere, the mortar constituents themselves were then examined in-situ with the naked eye and a 10x hand lens and described with reference to a classification scheme based upon the various aggregates and any carbonate/lime burning relicts displayed (Thacker, in press b). The lack of previous mortar archaeology in Atlantic Scotland initially required that this project take a very broad approach, and much work remains to be done. However, based on the emerging recorded evidence, we can now draw a number of preliminary conclusions: The earliest widespread evidence for lime mortar in Atlantic Scotland appears in chapel buildings of broadly 12th-century construction. There is some, much rarer, lime-mortar evidence in earlier unicameral chapel buildings, but these structures either demonstrate possible secondary lime contexts (as demonstrated at the Brough of Deerness [Morris and Emery 1986]) or are only relatively dateable as earlier phases of chapel structures that eventually developed a bicameral form. The earliest widespread lime-mortar evidence appears coeval with the building of bicameral chapels with a narrower chancel and, following Fleming and Woolf (1992), this plan-form has a coherent contemporary distribution across the Western Isles, Northern Isles, and Caithness (Fig. 1). Much more evidence for this distribution is being collated for this project, demonstrating a very concentrated archaeological record. Geographically, this pattern also conforms closely to distributions of “Papar” placenames (Crawford 2005), and finds of various examples of late Norse material culture such as “fish-tail” combs (Clarke and Heald 2002). In a century characterized in Scotland by constructional diversity (Fernie 1986), these findings support an interpretation that this plan-form is a Norse cultural expression of identity. It is, of course, dangerous to argue from negative evidence, but there does appear to be a lack of evidence for this plan-form in the rest of Atlantic Scotland, including large areas south of northern Caithness, the whole Highland west coast, and in the Hebrides south of South Uist—such as at Skye, Islay, Iona, and most of Argyll despite good preservation of other contemporary chapel plan-forms here. The use of lime mortar in both bicameral and earlier unicameral chapels is broadly contemporary with the use of other recorded bonding types, including dry-stone and clay (see Appendix 1). Mortar, whether lime or clay, would not then appear to be a structural requirement of the plan. Furthermore, 12th-century remains suggest a strong element of regionality within the province, with coherent Journal of the North Atlantic 48 M. Thacker 2015 Special Volume 9 Figure 1. Distribution of evidence for simple 12th-century bicameral chapels with a narrower chancel. From Survey 1 and Appendix 1. Journal of the North Atlantic M. Thacker 2015 Special Volume 9 49 evidence for lime use in Orkney, a more mixed picture of both lime use and survival of evidence in the Western Isles and Shetland, and a coherent record for the use of dry-stone in Caithness. However, suitable limestone for lime making is available in Caithness, e.g., at Raey, Baligill, and Durness (Saxon 2006), where dry-stone chapels are the norm in this period, and was apparently unknown in the Western Isles where lime bonding is evident. Accepting equality in the availability of fuel, this suggests the decision of whether to build with lime, clay, or dry-stone was not simply predicated on the physical availability or constraint of carbonate materials. Surely an empirical knowledge of lime-burning technology would by this time have been widespread, if not necessarily lime-burning techniques specific to locally available materials (see below). In the bicameral chapels of the province, evidence for chancel arches and/or barrel vaulting is implicit where lime bonding is recorded (Figs. 2, 3). Chancel arches or barrel vaulting are unknown in chapels recorded as dry-stone3 where trabeate (lintel-headed) portals are ubiquitous (Fig. 4). Within the recorded evidence for simple bicameral chapels, barrel vaulting is only found in those examples from Orkney. Buildings described as dry-stone tend to have much thicker walls, relative to those that are described as lime-bonded. Although the inclining of window or door jambs is found in both later and earlier chapel buildings, this is a very common feature of 12th-century bicameral chapels of all bonding types, both arched and trabeate. Throughout the province, until the 19th century, internal and external lime coatings are implicit with lime bonding. Further, surviving coating contexts, at least in the Western Isles, are generally contemporary with the underlying core masonry, of whatever phase. Pre-19th century re-pointing was not recorded, and even major structural developments during the life of a building did not generally result in recoating the whole structure. Further, no evidence has yet been found for the recoating of a chapel in its later role as a roofless burial aisle (hereafter cabeil), or, conversely, for the intentional early post-Reformation stripping of coatings. This evidence contrasts with that for the removal of both internal and external coatings, and sometimes re-pointing, during 19th- and 20th-century chapel restorations. Where unaffected by these late procedures, the archaeological record presents a series of remarkably pristine, if fragmentary, surviving pre-Reformation lime contexts that can serve as a valuable survey tool. Coatings are not recorded in almost all dry-built chapels. Clay-bonded chapels form a small percentage of both surveys, and evidence for chancel arches in these buildings is mixed. However, evidence for external lime coatings is not recorded, and only Eynhallow (Orkney) has any recorded external lime evidence at all. In contrast, the lower status and much later 19th/20th-century clay bonded “white- Figure 2. Lime-bonded, arcuate, Teampull Eoin, Lewis ca.1910–1913. Photograph © N. Morrison, donated by F. Macleod. Journal of the North Atlantic 50 M. Thacker 2015 Special Volume 9 houses” of Lewis usually had an external lime coating, to various extents. Lime mortars in Western Isles buildings generally display burnt relicts of limestone or sea shells visible to the naked eye, and this evidence is almost always mutually exclusive (Thacker 2012). A more specific terminology is therefore proposed, and will be used throughout this paper, to describe these two binder types as limestone-lime and shell-lime, respectively. A spectrum of aggregates from almost completely lithic to almost completely shell is also displayed—and so the four main mortar types will be defined as: limestone-lime/lithic tempered; limestone-lime/shell tempered; shell-lime/lithic tempered; and shell-lime/shell tempered (Ibid.). Shell-lime mortars, of both aggregate types, appear in chapel mortars of the Western Isles from the earliest period. This finding contrasts with contemporary evidence from mainland England, Scotland, and much of Ireland, where there is strong evidence for limestone-lime mortars, but does parallel some evidence from the 12th to 17th century in the Northern Isles (O’Dell 1959), late medieval Faeroes, Greenland, and Holland, and early modern evidence from Galloway (Scotland). The shell-lime relicts evidenced are overwhelmingly of cockle shell across all of the Western Isles except Lewis. Cockle shelllime mortars are occasionally found in Lewis also, but other pre-Reformation chapel mortars here appear to employ different shell types. Unfortunately, where medieval lime sources other than shell have been suggested, such as at Tur Chliamain (Harris) and Teampull Pheadair (Shader, Lewis), upstanding, non-invasive direct evidence is not available. However, lime-coating evidence which predates the conservation pointing at Tur Chliamain, suggests a shell-lime was probably also the primary core mortar material here. Further, Headrick’s “experiments” with burning marl on Lewis was a recent 18th-century innovation (see Anonymous 1791–1799 entry for Dunnichen:426–427) that was not being burnt by Lewis builders at the time. Reece’s (1981:15–16) suggestion that “white sand” was burnt to produce lime in medieval Iona is particularly interesting here but is an interpretation that may unfortunately have added to confusion over binder and aggregate sources, and further complicated mortar descriptions and classification (Thacker 2012). The exclusive use of shell-lime rather than limestone- lime mortars in the Western Isles is evident in Figure 3. Looking towards the chancel, lime-bonded, arcuate, an d vaulted Egilsay, Orkney. Photograph © M. Thacker Journal of the North Atlantic M. Thacker 2015 Special Volume 9 51 20th centuries, supports an interpretation for some form of unbroken technology, if not necessarily technique, in the southern Western Isles from the 12th century to the present day. During these later the archaeological record throughout the whole of the Late Norse and medieval periods. Physical evidence in later post medieval and post-Reformation buildings, and historical evidence from the 18th to Figure 4. Looking from the chancel at dry-stone St. Mary’ s, Caithness. Photograph © M. Thacker. Journal of the North Atlantic 52 M. Thacker 2015 Special Volume 9 this material as a “fine” lime (e.g., Wallace 1684). In this instance, then, a hot mix should not be ruled out on this basis. Across 12th-century Atlantic Scotland, the bicameral plan-form suggests shared identities of one order, while the constructional variability demonstrated above suggests diverse “accents” at a more intimate scale. The meanings of lime now need further examination if we are to interpret the reasons for the variability of this evidence, as each constructional solution is the result of a contextually specific, encultured, and socially negotiated practical process (Rapaport 1969). Technology and technique exist as interrelated yet separate (Ingold 2000, Inker 2000:26) and active media by which identities are negotiated. Perhaps, like Bede from our introduction, the bicameral chapel builders of Caithness and Orkney are expressing dialectics of regional identity, but the notion of group consensus may not be appropriate when all material culture is the result of the negotiation of power between agents at many different scales. We must push our explanations further and attempt to understand how decisions were orchestrated by social and material agents, with reference to implicated material and technological meanings. Were the techniques specifically associated with shell-lime also an expression of identity, and was the final plan form of the chapel less or more important to various agents than the techniques by which it was made? Discussion Lime-bonded, and so lime-coated, living castles and chapels would have had particular power in a treeless Western Isles landscape in which even the other buildings are composed of turf and stone. In order to understand more about lime, therefore, it is also necessary to investigate these other materials with which people needed to negotiate on a daily basis. Turf, stone, and lime are all powerfully evocative pieces of meaningful places (Bradley 2000:chapter 6), and the settlement landscape itself emerges from their interactions with each other and us. From a human perspective, these interactions are techniques for living—constantly reflexive processes from which the meanings of various people, materials, places, and in this case buildings, also emerge. In this way, the world emerges, through technique, from the landscapes of the past. Although, as buildings archaeologists, we are often more concerned with discreet dates, structures, and phases than the constantly shifting negotiation, the rite of dedication of a chapel building, like all other rites of passage, marks neither completion nor beginning; it marks contexts, shell-lime use was contemporary with the use of limestone-lime, and some buildings and sites display shell-lime evidence which overbuilds or post-dates limestone-lime contexts. No evidence for post-late-18th-century shell-lime use in Lewis, however, has yet been identified. As introduced above, historical evidence describes the importation of limestone to Lewis and Harris from the late 18th century, while this project demonstrates earlier, physical, 17th-century evidence for limestone-lime in Lewis. This early appearance may initially have been restricted to the Stornoway area—pre-figuring somewhat early cement evidence here in the mid-19th century. The lack of historical accounts of shell burning in Lewis parallels both this relatively early physical evidence of limestone-lime in some high-status lime-bonded buildings, and the later widespread evidence of clay-bonded, limestone- lime–coated white-houses on the island. This suggests that widespread shell-burning in Lewis, at least initially in the Stornoway parish, was replaced by limestone-lime and limestone importation at a much earlier date than in Harris, the Uists, and Barra. The late-18th-century historical shell-burning account from Uig (Lewis; Monro 1791–1799:42) is, however, later than any physical evidence known to this author, and much more mortar archaeology needs to be done in later contexts across the islands before we can suggest with greater authority when the technique might have been abandoned. Limestone-lime was, however, eventually active throughout the whole of the Western Isles—at the same time in different places to shell-lime, at different times to shell-lime in the same buildings, and there is even some later, rare evidence for different limestone- and shell- lime mortars in the same building at the same time. A local aggregate source was almost invariably used, in all periods. Within local arrays of possible aggregates, of very distinctive local composition, conscious choices appear to have been made, although occasionally these choices may conflict with a 21st-century mason’s appreciation of functionality. This point is important in confirming that the early mortars were mixed at or very near the chapel site, and enable further focused examination of carbonate relicts. As will be discussed below, however, this evidence also allows us to approach an understanding of the perceptions, experience, and meanings of the particular locales from which these materials were sourced. The lack of obvious lime inclusions in shell-lime mortars, which are so commonly found in later limestone-lime mortars, may be a consequence of the thin source material, and further, may explain why historical descriptions often seem to refer to Journal of the North Atlantic M. Thacker 2015 Special Volume 9 53 process and change (Bradley 1998). The social life of the building, like that of the people, materials, places, and techniques of which it is constructed, began long before this point (cf. Sinclair 1995). Emergent processes force us to consider whether, or how, meanings have changed even when the material record may display a marked degree of consistency. Much has been made of a Hebridean blackhouse building tradition, which, it has been suggested, has survived for the last 1000 years (Holden 2004, Smith et al. 2001,Walker and MacGregor 1996). Throughout much of this period, however, these houses were physically ephemeral turf structures that moved around the settlement landscape. 19th- and 20th-century material evidence, and tenant leases and fines, are testament to how the imposition of stone and lime “improvements” by various landlords throughout the Highlands and Islands were resisted by communities here. To contrast function with symbolism is unhelpful; this material culture both includes and represents much more than availability or husbandry needs (Dodgshon 1993:423). Turf walls and thatch roofs enabled incorporated building performances to continually re-affirm the communities’ relationships with the world through a constant cycle of turf, straw, and peat cutting; building, domestic fire rituals, and prayer; and finally the spreading of smoke-blackened turf or thatch upon the land. That stone and lime are very powerful agents with which to negotiate, which might slow this pulse to a standstill, was well understood (Thacker 2011). This incorporated, mobile ephemerality within the settlement landscape, however, also completely contrasted with the inscribed, static, oriented, monumental, stone and lime buildings with which this paper is primarily concerned. Western Isles pre-Reformation chapels are often sited upon the same ruined dry-stone buildings from earlier periods that Piggott interpreted as so culturally defining. Consistency of place here, however, has more recently been interpreted as the appropriation of a community’s pre-existing relationship with a powerful monument—often sited in liminal locations of commanding visibility (Driscoll 1998, Parker Pearson et al. 1996). Appropriation is one of the foremost tools of colonization and conquest, and like the earlier Iron age buildings, chapels returned a culturally constructed gaze very much dependent upon the relationship between the individual human being and building concerned (Olin 1996)—these are two negotiating, intersubjective “viewees”. No longer, if ever, simply a dry-stone broch, however, the sacred building place is now associated with and surmounted by Christianity, and has a lime-coated gaze which required those people who came within view to kneel or bow their head to pray (Martin [1695] 1970:28, 88). Although interpretations suggesting that masonry chapels were always lime-coated in medieval Europe have recently been questioned (Armi 1990), the physical lime-coating evidence apparent in the lime-bonded chapels of the above survey does parallel documentary and place-name evidence for white lime-coated or -washed chapels elsewhere in early medieval England (Cramp 2006) and Galloway (Hill 1997), 12th-century Wales (Gem 2009) and Scotland (MacDonald 1999:187), and medieval Greenland (Nyegaard 2009:7). Today, in many Christian contexts, perceptions of this color have powerful culturally loaded meanings which, in opposition to the color black, symbolize purity, peace, innocence, cleanliness, virginity, goodness, and life (Darvill 2002:74). In the last two centuries in the Hebrides, white has appeared as Chrisom cloths, confirmation dresses, wedding dresses, wedding flags, and Presbyterian Orduighean sheets (Parman 1990). From prehistory, however, the moon and pieces of quartz have had celestial and water associations in a Hebridean archaeological record that appears to further illustrate how an apparently universal and fundamental human relationship with this color has been articulated in different contexts (Jones 1999, Moore and Terry 1894). In the early medieval period, Coelfrid explained to the Pictish king Nechtan the pivotal role of the moon in calculating the correct date of Easter, and further how: “... we should celebrate the mysteries of our Lord’s Resurrection and our own deliverance with our minds refreshed to love of heavenly things ... the Lord Jesus, overcame all the darkness of death by the triumph of His Resurrection and then, having ascended into heaven, sent down the Spirit from on high and so filled His Church, which is often symbolically described as the moon, with the light of inward grace.” (Bede [731], Shirley-Price 1990:315; my italics). The moon, as God’s light, guides us through the night. In the Late Norse period, Christian meanings were constructed around white as the means by which God actively protects those in transition between two worlds. Again, the pre-existing cultural landscape, in this case the meanings surrounding a color, the night, and the heavens, appear to have been appropriated in a process of negotiation and power. White lime-coated chapels, like the saints whose relics they very often contained, loomed protectively over the people and landscape (cf. Airlie Journal of the North Atlantic 54 M. Thacker 2015 Special Volume 9 to fire a kiln that must also, in the pre-Reformation period, have been blessed (Thomas 1971:32). The source material was transformed by fire and its inherent sanctity revealed as “quick”, animated, and alive, with a thirst that required slaking. Water, milk, eggs, blood, and alcohol (Crhova et al. 2010, Sickels and Alsopp 2005), have all been involved as slaking materials elsewhere in Europe during the medieval period, and similar evidence is emerging as a result of more focused research here—blood in the mortar at Trondheim Cathedral, “liquor” at Dunvegan Castle, and blood, eggs, and even human hair at Scalloway Castle (Bjørken 1994, MacLeod 1906). These are visceral, powerfully emotive agents and bodily fluids necessary for life and so the safe birth of lime. The Late Norse or medieval lime-burner would not have worked by “scientific” knowledge of material properties, but by “enminded” technique to conduct a negotiation between material agents (Ingold 2000:chapter 16). Their performance was embodied in the resultant quicklime, which was then bound to place by mixing with very particular aggregates, sourced from the local shore. In the Western Isles, however, this locale was already very much implicated in the life and identity of the lime-burner; it was from here that the shell carbonate source had already been gathered. As introduced above, much of Atlantic Scotland presents a coastal landscape, and the influence of the sea on life here has often driven explanations for the apparent initial early medieval success of Norwegian colonization (Crawford 1987:11). In the Western Isles, the sea, like the wind, is an almost ever- present powerful agent that has probably always played a massive part in structuring personal experience (cf. Tilley 1996). The aggregate arrays within the historic Western Isles mortars surveyed in this research display very little change from that in evidence on the shore today. Modern science describes how these materials have been created from a mix of coralline algae from living offshore banks, stone and minerals from eroding onshore bedrock, and shell material from the fore-shore to form unique local compositions (Dawson et al. 2004). Collecting shell and aggregate for lime-making, however, demanded people enter this place of ancient and constant negotiation between land, sea, and sky (cf. Bradley 2000:136); the beach has always been a shifting amphitheatre of powerful synaesthetic experience. Particular beaches, however, also mean much more to different people, as places of experience, emotion, habitual bodily movements, and memory. The shore now appears to this writer replete with meanings associated with journeys and the charged emotions of leaving and arrival. It has a particularly “temporally 1994:36–37). The unique glow of lime coatings, especially at sunrise and sunset, may be explained optically by the “double refractive index” of the lime crystals themselves (Roz Artis-Young, Scottish Lime Centre Trust, Fife, Scotland, UK, pers. comm.), but “throughout the world brilliant things equate with spiritual enlightenment … and could have been conceived as earthbound material manifestations of light’”(Saunders 2002:214; my italics). Lime, quartz, and the moon are more than white— they glow, they are candindam4. In the 8th century, Bede compared Roman Catholic stone chapels with timber examples from elsewhere, but in many medieval Western Isles’ contexts, it is surely the external lime coatings that would have visually confirmed the loca sanctorum, and the presence of Celtic saints, the Roman Church, and God. Later in this period, the regional timing of which is also instructive, lime-bonded and -coated clan castles would also appropriate and contrive this physical mechanism—further linking the sanctity of lime with the clan chief. Lime-coated buildings are investment (cf. Gondek 2006) and performance opportunities, which reach back in time and up to heaven to monumentally present embodied, complex, hybrid relationships and identities within the landscape (Driscoll 1998; 2000:249). Lime mediates the cult of divine Christian lordship, and whether making swords, carving stones, or building chapels, the respective smith, carver, mason, and lime-burner are all political personnel (cf. Peregrine 1991:1). Lime-burning, however, like iron working (Haaland 2004), has often been studied by archaeologists from an implicitly modern perspective whereby technologies evolve and diffuse based upon rational criteria of efficiency and product performance (e.g., Leach 1995, Wingate 1985). It is, of course, useful to recognize how this purified, rational “spin” was adopted to explain different motivations in later periods—for instance, during the 19th-century blackhouse “improvements” discussed above—but it is also necessary to appreciate that other meanings are implicit here, and certainly would have pertained in 12th-century Atlantic Scotland. In whichever context, however, the lime-burner was one of many different agents implicated in a performance surrounded by creative and contingent symbolism. Limeburning required fire: a vital, sacred, and precious phenomenon prayed over in nightly post-medieval Hebridean domestic smooring rituals for centuries (Gavin- Schwartz 2001); it was not only the moon which kept the household safe. In the context of building large, thick-walled, lime-bonded structures, however, peat was demanded in enormous quantities Journal of the North Atlantic M. Thacker 2015 Special Volume 9 55 In these burials, Gilchrist (2008) discusses the apotropaic quality of shell amulets as the pilgrim undertakes the final journey through purgatory, but further, the archaeological record of early Christian burial across Britain documents bodies laid not in soil but in shell sand, chalk, lime, or even exotic Gypsum (e.g., Sparey-Green 2003). This process is also seen in Pictish Sutherland (Brady et al. 2007), and is strikingly recounted in the burial of Aud the Deep-Minded who was purported to have been buried “on an Icelandic beach at the high-water mark because having been baptised, she didn’t wish to lie in unconsecrated earth” (Pálsson and Edwards 1978:51–52 and 55, cited in Abrams 2007:183). Again in direct contrast to the inland cutting of turf for the incorporated ephemeral domestic buildings of this whole period, monumental chapel buildings display their shoreline origins in lithic and shellmortar aggregates, binder relicts, and water-worn building stones. These are powerful, universal but vernacular agents. Later medieval chapels embraced a wider cultural landscape, hybridizing with Mull/ Ionan Carsaig sandstone details and Argyll Schist graveslabs, but appropriation isn’t confined to the modern concept of “cultural reuse” and it is problematic to define motivations based upon simple categories such as “casual”, “functional”, or “iconic” (contra Stocker and Everson 1990). Sourcing shell and aggregate from the shore for their transformation and rebirth into white shell-lime mortars also mediates an evocative interpenetration of place whereby the chapel is now associated with a powerful, meaningful locale, and the shore becomes a Christianized locus. The Western Isles pre-Reformation chapel is an extension of the shore which it very often overlooks—a place of transition, contrasting perspectives and so power. The apparent dyadic relationships (Tilley 2004:4–10) discussed above between white or black, shore or inland, and lime-plastered or not, may go some way to explaining the dichotomy between the evidence from 12th-century lime-bonded Orkney and dry-built Caithness as highlighted in Survey 1. That this evidence also parallels other regional variations in Norse material culture is very pertinent (Clarke and Heald 2002), but we might now go one step further and question the premise of a Scottish Atlantic, or Scandinavian Scottish, cultural milieu which was uniformly dominated by the sea. Caithness is an active landscape very different from that of Orkney and not only because it is a part of the Scottish mainland. The Northern Caithness coastline presents a physical bulwark of high cliffs, with only small intervening sections of sand and boulder beaches, and one island (SNH 2002:100). thick” quality—an immediacy of physical experience which somehow engenders reflection. In the 19th century, however, shellfish was often starvation food, and the shore was structuring a very different experience. In earlier centuries, the western trackless ocean, the Uist “an Cuan an Iar”, was the direction from which the inexplicable often emerged in Gaelic proverbs (Campbell [1927] 1968:17–18). The ritual ale sacrifice to the sea-god Shonny, poured into the sea by wading 17th-century Lewis folk in supplication for a good seaweed harvest, further describes this embodied landscape and personified ocean (Martin [1695] 1970). Earlier still, in 16th-century Barra, the huge quantity of cockle shells found on the beaches of Eoligarry were reported to emerge in embryonic form from the holy well at Cille Bharra (Munro 1961), with all the liminal, journeying symbolism such a provenance might infer. But what of the earlier medieval and Late Norse periods? What did the shore mean to the parents of a young clansman who, having watched him climb aboard a galley with bravado, apparently ready for war, were now waiting impatiently for his safe return? Within this region, like the color white, emotive liminal locales such as the moon, brochs, wells, and beaches are apparently universal powerful agents with which to negotiate in our everyday lives and strong metaphors for wider concepts of the journey and where we stand in relation to a larger other. In the religious flux of the early second millennium A.D., such powerful Scottish Atlantic agents, like their human counterparts, had to be appropriated and Christianized. Elsewhere, scallop and cockle shells were establishing an important and widespread place in Christian material culture as the motif above the door of Christ’s tomb in Jerusalem (Ó Carrogáin 2003:144) and, from the 12th century, pilgrim badges to the shrine of St. James in Santiago de Compostela (Alford 1957, Hohler 1957). The shell has a particular resonance here in representing the journeys of both the headless body of St. James, blown to Spain by angels in a boat made of stone, and that of the pilgrims themselves. Pierced shell badges also then appear in iconographic images of James and other disciples right across Christendom, including on the beautifully carved Carsaig stone of Alistair Macleod’s late medieval tomb at Tur Chliamain (Harris). Pierced Scallop shells are found amongst the translated bones of medieval Irish saints at Illaunloughan (Edwards 2002:240) and Killoluaig (Ó Carrogáin 2003:144) and in the Christian burial contexts of pilgrims at 13th-century Taum, Galway (Clyne 1990), Fishergate, York, and the 12th-century church of St. Nicholas in Aberdeen (Gilchrist 2008). Journal of the North Atlantic 56 M. Thacker 2015 Special Volume 9 Significant settlement has historically been concentrated in the long river valleys characteristic of the region, and arable land is more widespread and, compared to Orkney, Shetland, or the Western Isles, much less coastal (Ritchie and Mather 1970). These contrasting landscapes are evident in the archaeological record of Caithness and Orkney from the Neolithic Period when, although both communities were building the same chambered cairn monument type, in Orkney they were oriented to ensure their gaze was felt most acutely from the sea while in Caithness it was concentrated on the inland “dales” (Phillips 2003). This emplaced monumentality is also evidenced in our survey, whereby half of the bicameral chapels of Caithness from Survey 1 (Skinnet, Gavin’s Kirk and Clow) are situated many miles inland while those of Orkney, like many in the Western Isles, are classic Norse beachhead sites. The 12th-century dry-built bicameral chapels of Caithness, however, are not evidence of simple environmental constraint, poverty, or conservatism in the region. They emerge from a more complex negotiation between Norse landowners, the Scottish Church, masonry traditions, ancient monuments, and the landscape (Thacker 2011). Given the evidence for limestone-lime mortars in much of Europe and Scotland, and the apparent widespread use of shell-lime mortars in the Norse Scottish Atlantic, even in regions such as Shetland and Skye, which like Caithness do have significant limestone outcrops, we should not simply question if knowledge of limestone-burning techniques was available, but how culturally appropriate those materials and techniques were. More mortar archaeology fieldwork within these regions is certainly required. Further, however, the various bicameral architectural forms included within Survey 1 represent a range of dates within the broad chronological parameters set and will include chapels built both before and after the establishment of an effective parish system in different regions. The concentrated distribution and evidence that most were not parish churches, however, suggests many were founded as “private” or “proprietorial” churches by landowners (Abrams 2007, Cant 1984) who also maintained the priest. Bicameral chapels evidently were the culturally appropriate form in many instances— a norm for a landowner of this status to aspire to conform to. Christianity, in appropriating the preexisting local landscape, was itself being appropriated and localized. Building on this interpretation, the adoption of the bicameral plan-form would then appear to reflect the Christianized Norse identity of these people in the first instance and, given our understanding of the fiercely independent political position often taken by members of this landowning “class”, the majority were probably not initially intended as overt expressions of reform. This interpretation does not discount the possibility that some chapels might have developed this association in contexts where the clergy were more powerful, but, crucially, we do not yet know specifics about where different Norse church proprietors stood or sat for the mass, or when they entered the building relative to the priest—perspectives from within the body of the church. A non-reform interpretation for the majority of these bicameral chapels, however, is underscored by the lack of evidence for this plan-form in Argyll, where evidence for reform monasticism is very much stronger during this and later periods—evidence that is almost completely lacking further north in the Western Isles (Raven 2005:165, Thacker 2011:60–61). The sanctity of lime is demonstrated in one of the earliest contexts we have in this province where, at the pre-Romanesque unicameral chapel on the Brough of Deerness in Orkney, lime mortar was used in a secondary phase to plaster the building’s internal east-end and altar only (Morris with Emery 1986). Like bicameral plan-forms, however, lime was also not initially an expression of reform in many contexts but as discussed above, paradoxically often an expression of the closeness of the lay/ Church relationship. This interpretation may help us understand something more of the Caithnessian dry-stone chapels of Survey 1 given our historical understanding of the very different relationships 12th-century Norse landowners had with the Norse Church in Orkney, and the Scottish Church in Caithness. Historically low 13th-century tithe revenues from Caithness do not necessarily infer secular poverty and neither, archaeologically, do dry-stone chapels. Both sources of evidence may further demonstrate the: “... process of implementing Scottish policies and Scottish ecclesiastical usage in an area where church structures had already been developed according to Norse patterns of Christian society.” (Crawford 1989:130). The dry-built chapels of Caithness positively referenced a Norse Caithnessian secular inland cultural and political Christian landscape. Lime, however, like the Scottish Church was a relentless and powerful agent with which to negotiate and would eventually even become a part of Caithnessian networks. Bicameral chapels, of whatever structure, externally displayed the important nave/chancel distinction, and so presented a constant mnemonic for the performances that took place Journal of the North Atlantic M. Thacker 2015 Special Volume 9 57 inside, and although we might question where the proprietal landowners may have initially placed themselves within this dichotomy, the chancel was certainly occupied by the priest. In most of northern Atlantic Scotland, lime mortar had hybridized with this bicameral chapel form and so enabled internal plaster, chancel arches, and/or even chancel vaulting. Compared to the very low lintel-headed doorways set within dry-stone thick walls (for instance at St. Marys Crosskirk or St. Ronans North Rona), lime mortar had enabled architectural forms which physically dematerialized the nave/chancel boundary and yet negotiated a much more sophisticated separation—a distinctly synaesthetic performance of color, light, and music (Thacker 2011). External lime coatings here provided a new mnemonic for a more intense internal chapel experience—a powerful demonstration of another world that, rather like the experience of standing on the shore, mediated a perspective articulating the congregation’s relatedness to a larger divine other. Given these different architectural forms, it is important to highlight that in Survey 1 no attempt was made to refine the mostly late 11th-century to 13th-century chronology. This paper takes an inherently broad chronological approach in an attempt to describe a period of great change in terms of processes and meanings negotiated over the longer term (cf. Romankiewicz 2009). Nor was any attempt made at this stage to physically verify the evidence recorded from outside the Western Isles. In that respect, this initial stage in the research was more about investigating our current state of knowledge to set the agenda before continuing, broadening, and refining the work (see especially Thacker, in prep a). It is beyond the scope of this paper to discuss in detail the emerging evidence of changes in medieval masonry or how lime techniques developed at particular sites or from later contexts. Where sites were surveyed within the Western Isles, however, then the potential for on-site mortar archaeology to establish the development of specific buildings, and the wider site, was often very effectively demonstrated. Eaglais na h’Aoidhe (Lewis) for example, displays seven clear stratigraphic phases of contrasting shell and limestone mortars within the chapel walls (Knott and Thacker 2011; Thacker 2011, 2012, in press a), evidence which is allowing a more refined interpretation of other chapels in the region such as Teampull Aulaidh (Gress) and Teampull Colm Cille (Lochs) (Thacker 2011); Tur Chliamain (Harris) displays similar evidence over the wider chapel/ cabeil complex, and Cille Bharra and Kisimul castle (Barra) demonstrate relationships between secular and ecclesiastical buildings which have also helped to refine the development of both groups (Ibid.). Unfortunately, however, at Cille Donnain, only fragmentary basal courses remain—this was ultimately a dry-built interpretation based on a lack of mortar evidence, rather than definitive evidence of the absence of mortar (Fig. 5). Teampall Eoin in Bragar (Lewis; Fig. 2), however, as a substantially upstanding, bicameral, probably single-phase, 12th-century, shell-lime-bonded, internally and externally coated, chancel-arched chapel in the Western isles, provided a significant link between the two surveys undertaken above, and as a result, is emerging from this research as an even more significant site. Many aspects of the wider settlement, however, have already been well documented and now provide a valuable record of how negotiations between people, lime, turf, stone, sea, shore, and landscape physically emerged here (see also Barrowman 2005:32; 2008:15–17, 24–29). The chapel is sited on Iron age remains (ibid:15), but of further pertinence to the discussion above is research that suggests that, while Improvement was insisting on more monumental and so less frequent domestic building performances on individual crofts, in Bragar as elsewhere, this siting at this location was also a move inland—away from the sea, away from the shore, away from the lime-bonded chapel and its surrounding graves, and perhaps also away from the past (Thacker, in press b). But then a significant amount of the year had always been spent inland— “out the way”, away from the white glowing gaze which demands the kneel or bow, to the summer, the black lands, and the freedom of the sheilings. Conclusion The world emerges from negotiations between living people and living things from the landscapes of the past. The bicameral chapel emerged as an active agent within that negotiation in the archaeological record of many different cultural landscapes across Europe. These chapels developed a plan-form that architecturally articulated apparently universal human perspectives on another world through their use of space and the materiality of this world. The generic technologies associated with lime mortar suggest it was an agent for sacred universality: plastering over different geologies to create a glowing white specifically Roman gaze which, like the priest who performs within, might be found right across Europe; and, indeed, lime was soon to be active right across Atlantic Scotland—an agent for conquest and a powerful new negotiator for the “new” religion. But this universal lime technology was performed within variously emerging landscapes, by different Journal of the North Atlantic 58 M. Thacker 2015 Special Volume 9 In English archaeology, much has been made of the early medieval re-use of Roman sites and stone. Elsewhere I have argued that the dry-stone chapel masonry of 12th-century Caithness referenced the monumental building techniques of an earlier age, as well as appropriating their sites—more “Pictitas” than Romanitas (Thacker 2011:58). This is questionable. As this research project continues, it will be a key objective to refine the chronology of these masonry buildings, question the development of these masonry techniques, and reassess external influences (especially from Ireland), but also, following Raymond Lamb (1989:269), assess “for how long and to what extent those [Pictish] institutions influenced the development of the Norse culture here”—continuous or otherwise. In this regard, and returning to the introduction of this paper, it may appear from the above surveys that universal “Catholicity” in Atlantic Scotland did not require monumental masonry architecture per se, but somehow Romanized monumental masonry—Roman arches, Roman vaults, and Roman plasters all enabled by lime mortars. This interpretation, however, is both problematic and full of potential when we realize that of the substantially contextually specific techniques, involving negotiations between diverse meaningful agents. In the various contexts of Atlantic Scotland, as elsewhere, these agents are powerful, culturally constructed hybrids with networked meanings associated with the various pre-existing cultural landscapes—materials (turf, shell, stone), places (brochs, beaches, the moor), people (lime-burners, priests, chiefs) and times (the Iron Age, last year, summer). The meanings of these agents were appropriated and Christianized— and a Christianized world emerged. Other cultural factors and other layers of identity were also at play, however, negotiating appropriate forms of Christian expression. The above surveys and the continuing research are beginning to demonstrate something of that process: how the meanings of some of these hybrids have emerged, and how some meanings subsequently changed even when the form in some ways remained constant. There were also, of course, new forms, new materials emerging from different places, and new negotiations—techniques which became, and continue to become, appropriate in the variously changing cultural circumstances displayed in the archaeological record. Figure 5. Cille Donnain 2010, from the northwest. Photograph © M. Thacker. Journal of the North Atlantic M. Thacker 2015 Special Volume 9 59 Bjørken, A. 1994. Restoring Kristiansten Fort to its 1684 condition. Gemini – Norwegian University of Science and Technology, Trondheim, Sweden. Available online at http://www.ntnu.no/gemini/1994-02E/sog_side_40. html. Accessed 21 March 2011 Bradley, R. 1998. The Significance of Monuments. Routledge, London, UK. Bradley, R. 2000. An Archaeology of Natural Places. Routledge, London, UK. Brady, K., O. Lelong, and C. Batey. 2007. A Pictish burial and Late Norse/Medieval settlement at Sangobeg, Durness, Sutherland. Scottish Archaeological Journal 29(1):51–82. British Geological Survey. 1992. Geology of the Outer Hebrides. HMSO, London, UK. Caldwell, D., M. Hall, and C. Wilkinson. 2009) The Lewis hoard of gaming pieces: A reexamination of their context, meanings, discovery, and manufacture. Medieval Archaeology 53:155–203. Campbell, D. [1927] 1968. Gaelic proverbs. Transactions of the Gaelic Society of Inverness 45:1–32. Cant, R. 1975. The Medieval Churches and Chapels of Shetland. Shetland Archaeological and Historical Society, Lerwick, Shetland, UK. Cant, R. 1984. Norse influences in the or ganisation of the medieval church in the Western Isles. Northern Studies 21:1–14. Clarke, D., and A. Heald. 2002. Beyond typology: Combs, economics, symbolism, and regional identity in Late Norse Scotland. Norwegian Archaeological Review 35:81–93. Clyne, M. 1990. A Medieval pilgrim: From Taum to Santiago De Compostela. Archaeology Ireland 4:21–22. Cramp, R. 2006. Wearmouth and Jarrow, Monastic Sites. Vol. 2. English Heritage, London, UK. Crawford, B. 1987. Scandinavian Scotland. Leicester University Press, Leicester, UK. Crawford, B. 1989. Norse earls and Scottish bishops in Caithness. Pp. 129–147, In C. Batey, J. Jesch, and C. Morris (Eds.). The Viking Age in Caithness, Orkney, and the North Atlantic. Edinburgh University Press, Edinburgh, UK. Crawford, B. 2005. The Papar Project: Inception, parameters, and purpose. Available online at http://www. paparproject.org.uk. Accessed 06 October 2010. Crhova, M., S. Kuckova, R. Hynek, and M. Kodicek 2010. Proteomic identification of milk proteins in historical mortars. Pp. 91–98, In J.Valek, C. Groot, and J.J. Hughes (Eds.). Proceedings of the 2nd Historic Mortars Conference and Rilem TC 203-RHM. Repair Mortars for Historic Masonry, Final Workshop, Prague 22–24 September 2010. Institute of Theoretical and Applied Mechanics of the Academy of Sciences of the Czech Republic, Prague, Czech Republic. Darvill, T. 2002. White on blond: Quartz pebbles and the use of quartz at Neolithic monuments in the Isle of Man and beyond. Pp. 73–92, In A. Jones and G. Mac- Gregor (Eds.). Colouring the Past: The Significance of Colour in Archaeological Research. Berg, Oxford, UK, and New York, NY, USA. upstanding dry-stone architectural forms recorded above as trabeate, all had either inclined jambs, corbelled opes, or vaults, and these are all arch forms too. Further, many lime-bonded contexts share these features also. It is time to bear the gaze of these hybrids again, and take another look. Acknowledgments I would like to convey my thanks to those who organized the conference in Uist, and to Rachel Barrowman and the two anonymous reviewers whose comments and suggestions markedly improved this paper. The results presented here were previously presented in an unpublished MA (buildings arch) thesis at York University (Thacker 2011), and the subject is now being pursued more widely in an ongoing PhD (arch) at the departments of Archaeology and Geosciences, University of Edinburgh (Thacker, in prep a). Literature Cited Abrams, L. 2007. Conversion and the church in the Hebrides in the Viking Age: “A very difficult thing indeed”. Pp. 169–194, In B. Ballin Smith, S. Taylor, and G. Williams (Eds.). West over Sea: Studies in Scandinavian Sea-Borne Expansion and Settlement Before 1300. Brill NV, Leiden, The Netherlands. Airlie, S. 1994. A view from Maastricht. Pp. 33–46, In B. Crawford (Ed.). Scotland in Dark Age Europe. The Committee for Dark Age Studies University of St. Andrews, St. Andrews, UK. Alford, V. 1957. The feast of Santiago in Galicia 1956. Folklore 68:489–495. Anonymous. 1791–1799. Entry for Dunnichen, County of Forfar. Available online at http://www.monikie.org.uk/ statisticalaccounts. Accessed 21 August 2010. Armi, C. 1990. Report on the destruction of Romanesque architecture in Burgundy. Journal of the Society of Architectural Historians 55(3):300–327. Armit, I. 1991. The Atlantic Scottish Iron Age: Five levels of chronology. Proceedings of the Society of Antiquaries of Scotland 121:181–214. Armit, I. 1996. The Archaeology of Skye and the Western Isles. Edinburgh University Press, Edinburgh, UK. Baden-Powell, D. 1938. On the glacial and interglacial marine beds of northern Lewis. Geological Magazine 75:395–409. Barnwell, P. 2004. The Laity, the clergy, and the divine presence: The use of space in smaller churches of the eleventh and twelfth centuries. Journal of the British Archaeological Association 157:41–60. Barrowman, R. 2005. Lewis coastal chapel-sites survey 2004/2005. Unpublished Document, Glasgow University, Glasgow, UK. Barrowman, R. 2008. Lewis coastal chapel-cites survey 2007–2008. Unpublished Document, Guard 2395, Glasgow University, Glasgow, UK. Bede [731] 1990. Ecclesiastical History of the English People. L. Shirley-Price (Trans.). Penguin, London, UK. Journal of the North Atlantic 60 M. Thacker 2015 Special Volume 9 Hingley, R. 1992. Society in Scotland from 700 BC to AD 200. Proceedings of the Society of Antiquaries of Scotland 122:7–53. Hohler, C. 1957. The badge of St. James. Pp. 49–70, In I. Cox (Ed.). The Scallop: Studies of a Shell and its Influences on Mankind. Shell Transport and Trading Co, London, UK. Holden, T. 2004. The Blackhouses of Arnol. Historic Scotland, Edinburgh, UK. Hughes, J.,and S. Cuthbert. 2000. The petrography and microstructure of medieval lime mortars from the west of Scotland: Implications for the formulation of repair and replacement mortars. Materials and Structures 33:594–600. Ingold, T. 2000. The Perception of the Environment: Essays in Livelihood, Dwelling, and Skill. Routledge, London, UK, and New York, NY, USA. Inker, P. 2000. Technology as active material culture: The Quoit brooch style. Medieval Archaeology 44:25–52. Johnstone, G.S., and W. Mykura. 1989. British Regional Geology: The Northern Highlands of Scotland. HMSO, London, UK. Jones, A. 1999. Local colour: Megalithic architecture and colour symbolism in Neolithic Arran. Oxford Journal of Archaeology 18:339–350. Knott, C.M., and M. Thacker. 2011. Eaglais na h’Aoidhe. Archaeological Survey. Unpublished document. Prepared for Urras Eaglais na h’Aoidhe and Historic Scotland, Edinburgh, UK. Knox, J. [1786] 1975. A Tour through the Highlands of Scotland and the Hebride Isles in 1786. James Thin, Edinburgh, UK. Lamb, D. 2008. Peeling back the layers: Reconstructing a vanished Iron Age landscape. Pp. 73–94, In G. Noble, T. Poller, J. Raven, and L. Verril (Eds.) Scottish Odysseys: The Archaeology of Islands. Tempus, Stroud, UK. Lamb, R. 1989. Carolingian Orkney and its transformation. Pp. 260–271, In C. Batey, J. Jesch, and C.D. Morris (Eds.). The Viking Age in Caithness, Orkney, and the North Atlantic. Selected papers from the proceedings of the 11th Viking Congress. Edinburgh University Press, Edinburgh, UK. Leach, J. 1995. Burning lime in Derbyshire pye kilns. Industrial Archaeology Review 17:145–158. MacDonald, A. 1999. Coming in from the margins: The Descendents of Somerled and cultural accommodation in the Hebrides, 1164–1317. Pp. 179–198, In B. Smith (Ed.). Britain and Ireland 900–1300: Insular Responses to Medieval European Change. Cambridge University Press, Cambridge, UK. MacIntyre, M. 1993. Making hydrated lime from seashells: As recalled by Domhnall lain Monk, Benbecula recorded December 1992 by Martin Maclntyre. Transcript published in ComunnEachdraidh nan Eilean mu Dheas 4:8. MacKenzie, Rev. C. 1791–1799. Statistical account of Scotland 1791–1799, volume 19, Parish of Stornoway county of Ross and Cromarty. Pp. 241–262. Available online at http://www.edina.ac.uk. Dawson, S., D. Smith, and J. Jordon. 2004. Late Holocene coastal sand movements in the Outer Hebrides, NW Scotland. Marine Geology 210:281–306. Dodgshon, R. 1993. West Highland and Hebridean settlement prior to crofting and the Clearances. Proceedings of the Society of Antiquaries of Scotland 123:419–438. Driscoll, S. 1998. Picts and prehistory: Cultural resource management in Early Medieval Scotland. World Archaeology 30:142–158. Driscoll, S. 2000. Christian monumental sculpture and ethnic expression in early Scotland. Pp. 233–252, In W. Frazer and A. Tyrrell (Eds.). Social Identity in Early Medieval Britain. Leicester University Press, London, UK. Edwards, N. 2002. Celtic saints and Early Medieval archaeology. Pp. 225-266, In A. Thacker and R. Sharpe (Eds.). Local Saints and Local Churches in the Early Medieval West. Oxford University Press, Oxford, UK. Etheridge, R. 1876. Note on the fossils from the glacial deposits of the northwest coast of the Island of Lewis, Outer Hebrides. Geological Magazine 13:552–555. Fernie, E. 1986. Early church architecture in Scotland. Proceedings of the Society of Antiquiries of Scotland 116:393–411. Fleming, A., and A. Woolf. 1992. CilleDonnain: A late Norse church in South Uist. Proceedings of the Society of Antiquaries of Scotland. 122:329–350. Gavin-Schwartz, A. 2001. Archaeology and folklore of material culture, ritual, and everyday life. International Journal of Historical Archaeology 5(4):263–280. Gem, R. 1983. Towards an iconography of Anglo-Saxon architecture. Journal of the Warburg and Courtauld Institutes 46:1–18. Gem, R. 2009. Gruffudd ap Cynan and the Romanesque church of Penmon, Anglesey. Pp. 301–314, In N. Edwards (Ed.). The Archaeology of the Early Medieval Celtic Churches. Maney, Leeds, UK. Gilchrist, R. 2008. Magic for the dead? The Archaeology of magic in later medieval burials. Medieval Archaeology 52:119–159. Gondek, M. 2006. Investing in sculpture: Power in Early Historic Scotland. Medieval Archaeology 50:105–142. Goodenough, K., and A. Finlayson. 2006. The Geodiversity of Harris: Statement of Significance and Identification of Opportunities. British Geological Survey, Nottingham, UK. Haaland, R. 2004. Technology, transformation, and symbolism: Ethnographic perspectives on European iron working. Norwegian Archaeological Review 37:1–19. Harding, D.W. 2000. The Hebridean Iron Age: Twenty years research. Occasional Papers Series no. 20, University of Edinburgh, Edinburgh, UK. Headrick, J. [1800] 1870. Report on the Isle of Lewis by the Rev. Mr. Headrick. Contained in a letter to The Right Hon. Lord Seaforth, The Proprietor. Vacher and Sons, London, UK. Hill, P. 1997. Whithorn and St. Ninian: The Excavation of a Monastic Town 1984–1991. Sutton/Whithorn Trust, Stroud, UK. Journal of the North Atlantic M. Thacker 2015 Special Volume 9 61 Pálsson, H., and P. Edwards, 1978. Orkneyinga Saga. The History of the Earls of Orkney. Penguin, Middlesex, UK. Parker Pearson, M., N. Sharples, and J. Mulville. 1996. Brochs and Iron Age society. Antiquity 70:57–68. Parker Pearson, M., N. Sharples, and J. Symonds. 2004. South Uist: Archaeology and History of a Hebridean Island. Oxbow, Oxford, UK. Parman, S. 1990. Orduighean: A dominant symbol in the Free Church of the Scottish Highlands. American Anthropologist 92:295–305. Parsons, D. 1990. Review and prospect: The stone industry in Roman, Anglo-Saxon, and Medieval England. Pp.1-15, In D. Parsons (Ed.). Stone Quarrying and Building in England AD 43–1525. Phillimore and Co., Sussex, UK. Parsons, D. 1994 The Church and its architecture before and after the Reformation. Pp. 264–282, In B. Vyner (Ed.). Building on the Past. The Royal Archaeological Institute, London, UK. Pearson, A. 2006. The Work of Giants: Stone and Quarrying in Roman Britain. Tempus, Stroud, UK. Peregrine, P. 1991. Some political aspects of craft specialization. World Archaeology 23:1–11. Phillips, T. 2003. Seascapes and landscapes in Orkney and northern Scotland. World Archaeology 35:371–384. Piggott, S. 1966. A scheme for the Scottish Iron Age. Pp. 1–15, In A.L.F. Rivet (Ed.) The Iron Age in Northern Britain. Edinburgh University Press, Edinburgh, UK. Rapaport, A. 1969. House, Form and Culture. Prentice Hall, Englewood Cliffs, USA. Raven, J. 2005. Medieval landscapes and lordship in South Uist. Ph.D. Thesis. University of Glasgow. Available online at http://www.theses.gla.ac.uk. Accessed 2 October 2010. Reece, R. 1981. Excavations at Iona 1964 to 1974. Institute of Archaeology Occasional Papers 5:15–17. Institute of Archaeology, London, UK. Ritchey, S. 2009. Rethinking the twelfth-century discovery of nature. Journal of Medieval and Early Modern Studies 39(2):225–255. Ritchie, W., and A. Mather. 1970. Beaches of Caithness: A survey of the beach, dune, and dune pasture areas of Caithness. Commissioned by the Countryside Commission for Scotland. Available online at http://www. snh.org.uk. Accessed 16 October 2010. Romankiewicz, T. 2009. Simple stones but complex constructions: Analysis of architectural developments in the Scottish Iron Age. World Archaeology 41(3):379–395. Saunders, N. 2002. The Colours of light: Materiality and chromatic cultures of the Americas. Pp. 209–226, In A.Jones, and G. MacGregor (Eds.). Colouring the Past: The Significance of Colour in Archaeological Research. Berg, Oxford, UK, and New York, NY, USA. Saxon, J. 2006. A geological tour of the North. Caithness Field Club Bulletin. Available online at http://www. caithness.org. Accessed 23 November 2010. Sickels, L., and P. Allsopp. 2005. Lime and its Place in the 21st Century: Combining Tradition, Innovation and Science in Building Preservation. International Building Lime Symposium 2005. NLA Building Lime Group, Arlington, VA, USA. Macleod, Rev. J. 1791–1799. Statistical account of Scotland 1791–1799, volume 10, Parish of Harris, County of Inverness. Pp. 342–392. Available online at http:// www.edina.ac.uk. Macleod, R.C. 1906. The Macleods: A Short Sketch of their Clan, History, Folklore, Tales, and Biographical Notices of Some Eminent Clansmen. The Clan Macleod Society, Edinburgh, UK. Macleod, R.H. 1998. Building St. Clement’s Church, Rodel, Harris. Halmac, Shetland, UK. MacQueen, Rev. E. 1791–1799. The Statistical Account of Scotland 1791–1799, Volume 20—Parish of Barray, The Western Isles. (Withrington and Grant [Eds.]1983). Malone, C. 2003. St. Bénigne in Dijon as Exemplum of Rodulf Glaber’s Metephoric “White Mantle”. Pp. 161–180, In N.Hiscock (Ed.). The White Mantle of Churches. International Medieval Research 10: Art History subseries 2. Brepols, Turnhout, Belgium. Martin, M. [1695] 1970. A Description of the Western Isles of Scotland. James Thin, EdinBurgh, UK Monro, Rev. H. 1791–1799. The Statistical Account of Scotland 1791–1799, Volume 20, The Western Isles. Parish of Uig. (Withrington and Grant [Eds.] 1983). Moore, A., and J. Terry. 1894. Water and well-worship in Man. Folklore 5:212–229. Morris, C., and N. Emery, 1986. The Chapel and enclosure on the Brough of Deerness, Orkney: Survey and excavations, 1975–1977. Proceedings of the Society of Antiquaries of Scotland 116:301–374. Munro, R. 1961. Monro’s Western Isles of Scotland and Genealogies of the Clans 1549. Oliver and Boyd, Edinburgh, UK. Myatt, L. 1975. The early ecclesiastical remains of Halkirk Parish. Available online at http://www.caithness. org. Accessed 24 July 2010. Nyegaard, G. 2009. Restoration of the Hvalsey Fjord church. In Norse Greenland: Selected Papers from the Hvalsey Conference 2008. Journal of the North Atlantic Special Volume 2:7–18. Ó’ Carragáin, T. 2003. The Architectural setting of the cult of relics in Early Medieval Ireland. Journal of the Royal Society of Antiquaries of Ireland 133:130–176. Ó’ Carragáin, T. 2005. Habitual masonry styles and the local organisation of church building in Early Medieval Ireland. Proceedings of the Royal Irish Academy 105C(3):99–149. Ó’ Carragáin,T. 2009. The architectural setting of the mass in Early Medieval Ireland. Medieval Archaeology 53:119–154. O’ Dell, A.C. 1959. Excavations at St. Ninian’s Isle. Scottish Geographical Journal 75:41–43. Olin, M. 1996. Gaze. Pp. 318–329, In R. Nelson, and R. Shiff (Eds.). Critical Terms for Art History. University of Chicago Press, Chicago, IL, USA. O’ Reilly, J. 2009. “All that Peter Stands For”: The Romanitas of the Codex Amiatinus reconsidered. Pp. 367–396, In J. Graham-Campbell, and M. Ryan (Eds.) Anglo-Saxon/Irish Relations Before the Vikings. Proceedings of the British Academy 157. Oxford University Press, Oxford, UK 462 pp. Journal of the North Atlantic 62 M. Thacker 2015 Special Volume 9 Withrington, D., and I. Grant (Eds.). 1983. The Statistical Account of Scotland 1791–1799, Vol. 20—The Western Isles. E.P. Publishing Ltd., Wakefield, UK. Young, A. 1955. Excavations at Dun Cuier, Isle of Barra, Outer Hebrides. Proceedings of the Society of Antiquaries of Scotland 89:290–327. Sinclair, A. 1995. The Technique as symbol in Late Glacial Europe. World Archaeology 27(1):50–62. Slade, H., and G. Watson. 1989. St. Peter’s Kirk, Thurso, Caithness c. 1150–1832. Proceedings of the Society of antiquities of Scotland 119:297–325. Smith, H., P. Marshall, and M. Parker-Pearson. 2001. Reconstructing house activity areas. Pp. 249–270, In U. Albarella (Ed.). Environmental Archaeology: Meaning and Purpose. Kluwer Academic Publishers, London, UK. Scottish Natural Heritage (SNH). 2002. Natural heritage futures: Orkney and Northern Caithness. Available online at http://www.snh.org.uk/futures/Data/pdfdocs/ Orkney_Caithness.pdf. Accessed 10 September 2010. Sparey-Green, C. 2003. Where are the Christians? Late Roman cemeteries in Britain. Pp. 93–108, In M. Carver (Ed.) The Cross Goes North. Processes of Conversion in Northern Europe, AD 300–1300. York Medieval Press, York, UK. Stocker, D., and P. Everson. 1990. Rubbish recycled: A study of the re-use of stone in Lincolnshire. Pp. 83–101, In D. Parsons (Ed.). Stone Quarrying and Building in England AD 43–1525. Phillimore and Co., Sussex, UK. Thacker, M. 2011. An archaeology of the lime and shelllime mortars of the Western Isles. M.A. Thesis. University of York, York, UK. Thacker, M. 2012. Towards an on-site mortar archaeology of the Isle of Lewis. Journal of the Building Limes Forum 19:38–44. Thacker, M. In press a. A Petrographic report on the microstructure of the lime mortars of the chapel-site at Eaglais na h’Aoidhe, Isle of Lewis. Historic Scotland, Edinburgh, UK. Thacker, M. In press b. The architectural development of post-Reformation burial aisles in the northern Hebrides. The Island Book Trust, Stornoway, UK. Thacker, M. In prep a. The quick and the dead: The archaeological potential of the lime mortars of the Mediebal and later North Atlantic. (working title). Ph.D. Thesis. University of Edinburgh, Edinburgh, UK. Thacker, M. In prep b. The ArdKenneth Lime Kiln Project: The surviving lime-burning heritage of the Western Isles. Thomas, K. 1971. Religion and the Decline of Magic. Penguin, London, UK. Tilley, C. 1996. The Power of rocks: Topography and monument construction on Bodmin Moor. World Archaeology 28:161–176. Tilley, C. 2004. The Materiality of Stone. Berg, Oxford , UK and New York, NY, USA. Walker, B., and C. Macgregor. 1996. The Hebridean Blackhouse. Historic Scotland, Edinburgh, UK. Wallace, J. 1684. Description of the Isles of Orkney. Available online at http://www.scran.org.uk Accessed 24 February 2011. Williams, R. 2004. Limekilns and Limeburning. Shire Publications, Princess Risborough, UK. Wingate, M. 1985. Small-Scale Lime-Burning: A Practical Introduction. Intermediate Technology Publications, London, UK. Journal of the North Atlantic M. Thacker 2015 Special Volume 9 63 Endnotes 1Very small isolated “limestone” outcrops have been mapped in the Western Isles, notably at Rodel, Harris (Goodenough and Finlayson 2006), and Garrabost, Lewis (Baden-Powel 1938). Given their size and rarity, however, it is perhaps unsurprising that these were often not apparently historically identified until either lime sourcing became an important Improvement strategy (e.g., Headrick [1800] 1870) or professional geologists studied the islands (Etheridge 1876). Macleod (1791–1799:372), for instance, states for Harris that “There is neither marble nor limestone nor freestone yet discovered”, while in Stornoway at this time “the houses are built at a considerable cost, because all the materials are imported, the stones not excepted …”. (MacKenzie 1791–1799:242– 243). The isolated carbonate-stone sources which have been recognized and documented will be important at some sites and do require more archaeological work (Thacker, in press a). At this point in the research, however, they should not detract from the interpretative and methodological (Thacker, in press b) consequences of a treeless Western Isles landscape which geologically consists “almost entirely” of quartzofeldspathic gneisses (British Geological Survey 1992, Johnstone and Mykura 1989:22). 2This masonry structure should be compared to the evidence for timber screens separating clergy from the congregation found in later churches and often presumed for earlier buildings. The chapel on the Brough of Deerness, however, did evidence a form of stone cancellum. For wider Insular discussion of how bicameral masonry mediates reform, see for example, Barnwell (2004), Ó’Carragáin (2009), Parsons (1994:278), or, more specifically for the Western Isles, Raven (2005:173–176) and Thacker (2011:40–47). 3St. Catherines, Linton, Orkney stands out here as the only chapel recorded as dry-stone with a chancel arch, thin chancel walls, and exterior pointing. These details require verification. 4Candida/candidus/candidam is usually translated as “shining white”, but the word has a much stronger spiritual resonance than this simple descriptive phrase might imply. It is found in medieval bible references to describe, for instance, Christ’s garments during the Transfiguration, the throne of God, the Eucharist, and a few other particularly significant contexts (Malone 2003:170). While there might appear to be some danger of tautology regarding use of this word to describe chapels, and particularly the plaster subsequently found at the putative site of Candida Casa (Whithorn), this point is irrelevant. Candidam is a literary metaphor for how we might perceive or conceive transition to the kingdom of heaven; importantly, the Scottish Atlantic archaeological evidence of shining white chapel lime coatings is evidence of that same metaphor made physical. For more discussion, especially with reference the apparent 12th-century “discovery of nature” (e.g., Ritchey 2009) see Thacker (in prep b). Journal of the North Atlantic 64 M. Thacker 2015 Special Volume 9 Appendix 1. 12th-century bicameral churches of the Scottish Atlantic Province.A No. refers to number as shown on map in Figure 1. Region/No. Chapel site NGR Bonding Wall width (ft.) Lime coatings Arching, vaulting, or inclined jambs Simple bicameral chapels with narrower chancel than nave Western. Isles 01. C. Donnain, S. Uist. NF 731 281 Dry 2’11”-3’0” No evidence Narrow (2’4”) chancel portal. 02. T. Eoin, Lewis. NB 288 489 Lime 2’5”–3’5 Interior and exterior coatings Chancel arch. 03. T. Pheadair, Lewis. NB 379 549 LimeB Turf-covered footings only. 04. T. Mhuir, N. Uist. NF 785 763 3’0 Turf-covered footings only. 05. T. Bhrìghid, Lewis. NB 409 573 Turf-covered footings only. 06. C. ‘Ic Ailean, S. Uist. NF 758 364 Lime 2’4”–2’7” In. and ex. coatings Chancel arch. Inclined jambs. 07. T. Ronain, N. Rona. HW 809 323 Dry 3’0–4’0” Internal coatings Lintelled portals. Corbel vault. Orkney 08. St. Mary’s, Wyre. HY 442 262 Lime 3’0” Interior and exterior coatings Chancel arch. ‘Harled’exterior. 09. St. Thomas, Rendall HY 424 210 Lime 3’0”–4’6” Barrel vaulted chancel. 10. St. Catherines, Lint. HY 529 018 Dry 2’10”–3’0 Exterior pointing Chancel arch. 11. St. Nicholas, P. Str. HY 669 291 Lime 2’0” Interior Plaster Chancel arch, Barrel vault. 12. Xkirk, Tuquoy. HY455 431 Lime Chancl.arch. Barrel valt. Incl. Jambs 13. Ladykirk, Westray. HY439 488 Clay 4’0” Chancel arch 14. St. Brides, Stronsay. HY666 291 Lime 2’0” Turf-covered footings only. 15. St. Peters, Evie. HY338 287 “Nave and chancel” Shetland 16. Uyea, Unst. HU 608 985 Lime/dryC 2’5”–3’3” Lintelled and corbelled chancel arch 17. Xkirk, Clibberswick. HP650 121 3’2”–5’6” 18. Kirk o Ness, N. Yell. HP532 048 LimeD 3’–3’9” Chancel arch 19. St. Mary’s, Sandstg. HU346 472 Lime 2’10” Chancel arch 20. St. John’s, Unst. HP651 141 2’4” Turf-covered footings only. 21. Meal Colvidale. HP622 045 3’0” Turf-covered footings only. 22. Kirkaby, Westing. HP566 064 2’9” Turf-covered footings only. 23. Hascosay, Yell. HU 545 918 2’4” Turf-covered footings only. 24. Kirkhouse, Fetglar. HU659 911 Turf-covered footings only. 25. Gungstie Noss. HU530 409 Turf-covered footings only. Caithness 26. St. Mary’s, Xkirk. ND 024 700 Dry 4’0” None Lintelled portals. Inclined jambs. 27. Skinnet, Halkirk. ND130 620 Dry/ClayE 3’6” None S. Entrance to nave and chancel. 28. St. Drostan’s. ND 317 693 Dry 4’0” Interior plaster 29. St. Mary’s, Clow ND 233 524 DryF 4’0” 30. Gavin’s kirk, Dorr. ND077 547 DryG 3’6” Turf-covered footings only. 31. Kirk o banks.D ND 253 739 Dry Turf-covered footings only. Argyll 32. St. Blane’s, Bute. NS 094 534 Lime Chancel arch. 33. St. Marnoc’s, Bute. NS 023 596 Lime 2’8” Chancel arch. Journal of the North Atlantic M. Thacker 2015 Special Volume 9 65 Region/No. Chapel site NGR Bonding Wall width (ft.) Lime coatings Arching, vaulting, or inclined jambs Complex Sites Orkney 34. St. Magnus, Egilsay. HY 466 303 Lime 3’0” Barrel-vaulted chancel. 35. Eynhallow. HY 359 288 Clay 2’6” Lime pointed Chancel arch. 36. Brough of Birsay. HY 247 277 37. St. Nicholas, Orphir. HY334 044 Lime Vaulted chancel. Shetland 38. St. Ninian’s. HU 368 209 Lime Interior and exterior coatings Chancel arch. 39. St. Mary’s Bressay. HU 521 422 Dry/Clay 2’3” Cruciform plan. Caithness 40. St. Peter’s, Thurso. ND 120 686 Lime 2’10”E Vaulted chancel. Western Isles 41. T. Moluaidh, Ness.H NB 519 651 Lime 2’9” Interior and exterior coatings Arched south entrance . Unicameral Western Isles 42. Cille Bharra, Barra. NF 705 073 Lime 2’6” Interior and exterior coatings Arches, inclnd jmbs, screen evidence.. Shetland 43. St. Olaf’s, Lund. HP566 040 Lime 3’6”–4’6” Inclined jambs. Argyll 44. Cille Dalton, Islay. NR 458 508 Lime 2’11” Interior coatings Wooden chancel screen. Oblong Bicameral Sites Western Isles 45. St. Brendon’s, BarraI NF 647 016 Lime 3’3” Orkney 46. St. Boniface HY488 527 Lime 3’0” Interior and exterior coatings Chancel arch. 47. Muckle House.J ND 325 992 Shetland 48. Chapel Knowe HU 485 691 Lime 2’0” Cross-wall, apsidal chancel. Caithness 49. St. Cuthbert’s, Wick.K ND 330 502 Argyll 50. Lismore Cathedral. NM 860 434 Lime 3’3” Later west tower. 51. KilChenzie. NR673 248 Lime 3’2” 52. St. John’s, Killean. NR 695 445 Lime 53. St. Cormac’s NR 666 752 Vaulted chancel. Skye 54. St. Assind’s NG 355 388 Lime 4’11” Journal of the North Atlantic 66 M. Thacker 2015 Special Volume 9 AThis list is taken from Thacker 2011. It is still, however, incomplete and currently being researched and compiled. It is presented here in interim. Details yet to be included include for instance those for the chapels at Portmahomack and St. Magnus, Birsay, and unpublished mortar archaeology details of upstanding monuments outside the Western Isles. BFrom oral tradition (Barrowman 2008:11). CRCHMS (although http://www.britishlistedbuildings.co.uk described as drystone. Accessed 24 July 2010). DSimilar to St. Mary’s Crosskirk, and similar to Gunstie Voe in plan (RCAHMS). EDescriptions vary. F“Walls apparently umortared”. GMyatt (1975). HWhile the foundation date of the church at Eoropie is uncertain and often contested, many do consider the church to be 12 th century, including Caldwell et al. (2009:176). ISt. Brendon’s is known to have had a nave and shorter chancel “constructionally seperated”. The lack of descriptions of the chancel width have led to an assumption here that it was oblong overall. JChancel described as “small and square” but here classified as i n footnote I above. KSlade and Watson (1989)