Insular networks: An exploration of material distribution, insularity and island identity on the edge of Europe in the Neolithic

Megarry, W. P., Markham, M., Gilhooley, B., O'Neill, B., & Cooney, G. (2021). Insular networks: An exploration of material distribution, insularity and island identity on the edge of Europe in the Shetland Neolithic. The Journal of Island and Coastal Archaeology. https://doi.org/10.1080/15564894.2021.1880506

Published in: The Journal of Island and Coastal Archaeology

Document Version: Publisher's PDF, also known as Version of record

Queen's University Belfast - Research Portal: Link to publication record in Queen's University Belfast Research Portal

Publisher rights Copyright 2021 the authors. Published with license by Taylor and Francis Group, LLC This is an open access article published under a Creative Commons Attribution-NonCommercial-NoDerivs License (https://creativecommons.org/licenses/by-nc-nd/4.0/), which permits distribution and reproduction for non-commercial purposes, provided the author and source are cited.

General rights Copyright for the publications made accessible via the Queen's University Belfast Research Portal is retained by the author(s) and / or other copyright owners and it is a condition of accessing these publications that users recognise and abide by the legal requirements associated with these rights.

Take down policy The Research Portal is Queen's institutional repository that provides access to Queen's research output. Every effort has been made to ensure that content in the Research Portal does not infringe any person's rights, or applicable UK laws. If you discover content in the Research Portal that you believe breaches copyright or violates any law, please contact [email protected].

Download date:07. Oct. 2021 The Journal of Island and Coastal Archaeology

ISSN: (Print) (Online) Journal homepage: https://www.tandfonline.com/loi/uica20

Insular networks: An exploration of material distribution, insularity and island identity on the edge of Europe in the Shetland Neolithic

William P. Megarry, Mik Markham, Bernard Gilhooly, Brendan O’Neill & Gabriel Cooney

To cite this article: William P. Megarry, Mik Markham, Bernard Gilhooly, Brendan O’Neill & Gabriel Cooney (2021): Insular networks: An exploration of material distribution, insularity and island identity on the edge of Europe in the Shetland Neolithic, The Journal of Island and Coastal Archaeology, DOI: 10.1080/15564894.2021.1880506 To link to this article: https://doi.org/10.1080/15564894.2021.1880506

© 2021 The Author(s). Published with license by Taylor and Francis Group, LLC

Published online: 30 Aug 2021.

Submit your article to this journal

View related articles

View Crossmark data

Full Terms & Conditions of access and use can be found at https://www.tandfonline.com/action/journalInformation?journalCode=uica20 THE JOURNAL OF ISLAND AND COASTAL ARCHAEOLOGY https://doi.org/10.1080/15564894.2021.1880506

Insular networks: An exploration of material distribution, insularity and island identity on the edge of Europe in the Shetland Neolithic

William P. Megarrya, Mik Markhamb, Bernard Gilhoolyc, Brendan O’Neilld, and Gabriel Cooneyd aSchool of Natural and Built Environment, Queen’s University Belfast, Belfast, UK; bImplement Petrology Group, London, UK; cIrish Antiquities Division, National Museum of Ireland, Dublin, Ireland; dSchool of Archaeology, University College Dublin, Dublin, Ireland

ABSTRACT ARTICLE HISTORY Shortly after arriving in the Shetland archipelago early in the 4th mil- Received 10 July 2020; lennium BC, communities began to quarry and make stone tools Accepted 4 December 2020 from riebeckite felsite, quarried from the region of KEYWORDS North Mainland. The effort expended traveling to the quarry sites, extracting, making and crafting tools was considerable indicating the Shetland Islands; stone tool distribution patterns; importance of felsite to Neolithic communities. Results from the geospatial modeling; Felsite Project are presented in this paper. The quarrying polished stone tools; and distribution of felsite in Shetland can tell us much about insular- insularity; Neolithic ity, choice, and island identities on a wider scale. The geospatial dis- tribution of 416 felsite tools from across the archipelago was explored. Felsite is used to make tools for utilitarian and ceremonial purposes and, with two exceptions compromising 0.33% of the total dataset, no tools can be securely provenanced outside of Shetland. At a time of increasing communications elsewhere in the Neolithic world, communities in Shetland appear to become focused on the archipelago. This insularity is also evident in contemporary material culture including pottery and funerary structures which do not fol- low mainland or Orcadian styles, strongly suggesting an island- focused society during this period. Recent radiocarbon dates suggest that these dynamics were established in the Early Neolithic, before 3500 BC. Felsite, therefore, allows us to explore the otherwise elusive early prehistory of the islands. The distribution network of tools from the quarry complex around the archipelago strongly indicates a pref- erence for maritime travel and transportation, even where terrestrial routes are available. The patterns also suggest settlement and activ- ity across Shetland with important foci at previously understudied locations across North and West Mainland. The study has consider- able implications for understanding movement, choice, and identity in past island societies and their relationship to material culture.

CONTACT William P. Megarry [email protected] School of Natural and Built Environment, Queen’s University Belfast, Belfast, BT7 1NN UK ß 2021 The Author(s). Published with license by Taylor and Francis Group, LLC This is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives License (http://creativecommons.org/licenses/by-nc-nd/4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited, and is not altered, transformed, or built upon in any way. 2 W. P. MEGARRY ET AL.

Introduction Background and context This paper will explore what stone tools can tell us about island society in Neolithic Shetland. While there is a broad discussion of all tools, the study is primarily focused on riebeckite felsite, a distinctive blue-gray rock from the Northmavine region of Mainland, which is found throughout the archipelago in this period (Cooney et al. 2019). For discussion in the paper the term felsite is generally used. Geospatial analysis and visualization techniques are used to explore the distribution of stone tools and to investigate transportation networks and stages of tool production and use across the archipelago. There is also a wider discussion on insularity and choice in Neolithic Shetland, exploring several issues relevant to this Special Edition, primarily the complex relationship between the environment and human agency. Specifically, it explores what the distribution of felsite tells us about social connectivity and choice within and out- with Shetland. It incorporates discussion on the paleoenvironment and terrestrial and maritime movement, using geospatial modeling to explore how people interacted with their landscapes in Neolithic Shetland. As one of the northernmost parts of Europe where agriculture was practiced during the Neolithic (Figure 1), the prehistory of Shetland remains somewhat of an enigma when compared to its southern neighbors, specifically Orkney, the neighboring archipel- ago (Edmonds 2019; Sheridan 2012). There are indirect paleoecological indicators, including potential grazing by large herbivores, for a sporadic human presence from around 5500 BC (Bennett et al. 1992), but the first definitive evidence comes from a Mesolithic-Neolithic transition site at West Voe, South Mainland at the start of the 4th millennium BC (Gillmore and Melton 2011; Melton 2008). Radiocarbon dates suggest that it was around 3600 BC when Neolithic communities became firmly established in Shetland (Cummings 2017; Sheridan 2012; Whittle 1986). Separated from mainland, Scotland by 160 km of open ocean (Figure 1), life on Neolithic Shetland was challeng- ing. Isotopes from juvenile dentin excavated at the site at Sumburgh, South Mainland, suggest periods of food shortage when terrestrial diets were supplemented by marine resources (Montgomery et al. 2013). Large-scale monumental architecture is also less prominent in Shetland than Orkney. Yet from early in the Neolithic, communities in Shetland were investing a significant amount of time and effort in one monumental- scale activity—the extraction and production of polished stone tools from felsite, quar- ried from the slopes of , the highest point of the archipelago and adjacent areas at North Roe, Northmavine, North Mainland (Figure 1). Felsite is a versatile lithic material (Reibeckite felsite is Group XXII in the British scheme of Grouping stone axes (Ritchie and Scott 1988)), and care seems to have gone into selecting felsite with dis- tinctive phenocrysts, spherulitic crystals, and/or banding. The extraction, manufacture and distribution of these tools has been the focus of the North Roe Felsite Project (NRFP), an ongoing multi-disciplinary research project established in 2012, focusing on the North Roe quarry sites and the 416 felsite stone tools from the archipelago (Ballin 2011; Cooney 2017; Cooney, Ballin, and Megarry 2012). Results from the project have fundamentally changed our understanding of the Neolithic in Shetland and provide important wider insights into the character of island communities in this period. THE JOURNAL OF ISLAND AND COASTAL ARCHAEOLOGY 3

Figure 1. Scotland, Shetland, and topography with key sites labeled (sites from Canmore). 4 W. P. MEGARRY ET AL.

Table 1. Radiocarbon dates from Neolithic Shetland. 14 Site Lab code Conventional age BP Calibrated C Date Source Modesty SUERC-37997 4580 ± 35 3500–3110 Sheridan (2012) Crooksetter SUERC-43349 4603 ± 27 3505–3360 Date provided by ORCA Crooksetter SUERC-43353 4582 ± 24 3494–3350 Date provided by ORCA Grut Wells UBA-35547 4580 ± 44 3379–3265 Dates from NRFP Grut Wells UBA-35548 4670 ± 33 3522–3366 Dates from NRFP Grut Wells UBA-35549 4650 ± 41 3522–3355 Dates from NRFP Calibrated using OxCal 4.1.7.

Neolithic Shetland: Settlement and chronology The locations of sites discussed are shown in Figure 1. Our understanding of settlement in Neolithic Shetland comes from a limited number of excavations and field surveys. Most of these, such as Calder’s study of house sites at Staneydale and the Ness of Gruting (Calder 1949, 1956), Whittle’s project at the Scord of Brouster (Whittle 1986), Mahler’s study of Pinhoulland (Mahler 2012), and Turner’s (Turner 2012; Turner et al. 2004) wider regional study of prehistoric field systems have focused on southern West Mainland. This has resulted in a strong regional bias which has impacted on our wider understanding of Neolithic Shetland. This situation is further exacerbated by a poor chronological framework for the period, which is overly reliant on predominantly Late Neolithic/Early Bronze Age dates from these excavations (Barcham 1978; Sheridan 2012, 2013; Whittle 1986). Yet, the assumption that West Mainland is an exemplar for the wider Shetland Neolithic from the Early Neolithic remains, to the detriment of our understanding of the period across the wider archipelago. Exceptions to this regional and chronological bias have emerged in recent decades, due largely to commercial excavations and projects like the NRFP. These projects have produced new radiocarbon dates from the previously elusive Early and Middle Neolithic periods (Table 1). Of significance is the association between some dates and polished felsite tools. Sheridan’s date (SUERC-37997) from a fagot of wood found by Kinghorn with a cache of felsite axes and knives in Modesty, West Mainland, in 1894 (Kinghorn 1894; Sheridan 2012) corresponds with three dates (UBA-35547, UBA-35548, and UBA- 35549) obtained by the NRFP from charcoal found in deposits related to the quarrying of a felsite dyke at Grut Wells and from an adjacent workshop (Cooney et al. 2019). Recent dates (SUERC-43349 and SUERC-43353) from the excavations at Hill of Crooksetter, North Mainland, further indicate that felsite exploitation (Cummings 2017), and the production of polished stone tools, was well established by the mid- fourth millennium BC. While further work needs to be done on the chronology of fel- site use, especially in the later Neolithic, current dates strongly suggest that felsite is a key indicator of Early and Middle Neolithic activity across the archipelago, and an important proxy for activity in this otherwise elusive period. Results from the NRFP suggest a complex archipelago-wide network of movement and distribution, shifting the focus away from West Mainland to other potentially important areas. This paper will explore this network, focusing on the distribution of felsite stone tools and what it can tell us about the manufacturing and distribution networks in Neolithic Shetland and wider regional communications. THE JOURNAL OF ISLAND AND COASTAL ARCHAEOLOGY 5

Materials and methods Felsite quarrying and tool production Based on artifacts in museum collections, particularly in Shetland Museum, Leriwick, and roughouts discovered during NRFP excavations at the Beorgs of Uyea and Grut Wells, a range of morphologically distinct felsite axes/adzes and knife types have been identified. Each of these tool types represents different strategies in the selec- tion and procurement of raw material: where to get desirable material; how to approach quarrying; extraction; and working strategies. Unmanned aerial vehicle (UAV) and GNSS survey and excavations at Grut Wells and at the Beorgs of Uyea have provided detailed information about the quarrying and production process (Cooney et al. 2019; Megarry et al. 2018). Although these quarry sites are different in their topography and scales of quarrying, the process of extraction and initial reduction is broadly comparable. In both locations, axes/adzes (the main tools pro- duced) were made from a petrological range of felsite, in dykes intruded into (Ronas Hill) granite. However, there appears to have been a preference for the finer-grained felsite found at the edges of dykes. Consequently, quarrying appears to have commenced alongside a dyke, removing granite to access more of its outer margins. Although there is currently no direct evidence for how massive felsite blocks were detached from dykes, it seems likely that a combination of percussion, topography, gravity, and wedging were employed. Unlike other Neolithic quarry sites across Northwest Europe, fire setting does not appear to have been used. Once removed the felsite blocks were reduced to roughouts in a series of stages. Surface survey at Grut Wells shows that large workable blocks were moved slightly downhill before being initially reduced or broken into smaller pieces. From this material blanks of different sizes were brought upslope for more formal roughing out. At the Beorgs of Uyea this pattern of staged mobility can also be seen with reduction largely occurring close to dykes and later stages identified further away (Cooney et al. 2019; Megarry et al. 2018). Examination of debitage and roughouts, as well as the abundance of hammer stones from these quarry sites and experimentation with this material, points to the use of hard-hammer direct percussion in all stages of manufacture around these quarry sites. Owing to the petrological variability of felsite, it is liable to hinge and step fracture while knapping, making grinding challenging and very time-consuming. Crucially, while there are final stage roughouts there is currently no convincing evidence for pecking, grinding, or polishing at or close to quarry sites. Instead, small/subtle indicators of these later stages have been found away from quarries, on sites often associated with habita- tion. While no grindstones have been identified to date, experimental work demon- strates that all stages of grinding can be carried out on the same medium-grained grindstone (Gilhooly 2018;O’Neill, Gilhooly, and Cooney 2014). The best evidence for polishing are small polissoirs, flakes or blades with discrete/distinct polishing surfaces, typically on their ventral faces at the proximal end (Figure 2), identified by NRFP ana- lysis. These had previously been interpreted by Ballin (2017) as recycled axeheads. This evidence for felsite tool production from different sites strongly suggests a spatially dis- persed cha^ıne operatoire on Neolithic Shetland. 6 W. P. MEGARRY ET AL.

Figure 2. Four polished feliste knives from various locations (top); a selection of felsite polissoirs from Punds showing dorsal (bottom left) and ventral faces (bottom right) (Photo: NRFP 2019).

Artifact catalog and geospatial analysis The NRFP has utilized geospatial technologies for a range of purposes including cata- loging artifacts and their provenance, global navigation satellite system (GNSS) for field survey and excavation, multispectral satellite data for prospection, and unmanned aerial vehicles (UAV) for the remote mapping of stone quarries and workshops (Megarry et al. 2016, 2018). This paper builds on this approach, exploring what the geospatial ana- lysis and visualization of the NRFP tools dataset can tell us about Neolithic society and felsite polished stone tool production on Shetland. Analysis was carried out in the open THE JOURNAL OF ISLAND AND COASTAL ARCHAEOLOGY 7

Figure 3. Treemap of artifact provenances (of all lithologies) from the NRFP (n ¼ 635). source Quantum GIS Program with added analytic functionalities provided by the GRASS toolkit. To date the NRFP has cataloged 658 artifacts to provide an overview of polished stone tools in Shetland (Markham 2018). Portable x-ray fluorescence (PXRF) has been used to identify lithologies for 80% of these artifacts, 416 of which have been identified as felsite. The remaining tools were made from other lithologies including gabbro, dior- ite, rhyolite, sandstone, amphibolite” and serpentines. As well as the use of secondary sources, some of these were quarried from primary sources across the archipelago. For example, there are sandstone quarries in West Mainland (Ballin and Mahler 2017). With regard to tool type the majority of these are axes (56%) and knives (37%). The remaining artifacts include battle-axes (1%) and cushion and pestle maceheads (2%). The remainder of this study will focus specifically on felsite axes and knives and no other lithologies or tool types. This subset consists of 395 artifacts. These axes and knives are primarily in museum collections in the UK, particularly in Scotland with a smaller number in local museums and private collections in Shetland. A large majority of artifacts are stray finds or come from antiquarian collections where the detailed provenance was not recorded (Figure 3). Of the cataloged felsite axes and knives, 128 (32%) have no recorded provenance while 267 (68%) can be provenanced at a parish scale. Detailed findspots are only recorded in a small number of cases, but broader find locations (towns, farms or crofts) within parishes are recorded for a further subset of 205 objects (52% of all artifacts). This provenance dataset was imported into a GIS and joined with relevant spatial datasets. The sourcing of felsite artifacts solely to the North Roe quarry complex makes them a unique dataset to explore distribution pat- terns across the archipelago. Of these objects, 201 are axes (51%) and 194 are knives 8 W. P. MEGARRY ET AL.

(49%). These artifacts show varying degrees of use. Ballin has made a division between prestige and “everyday” axeheads, and the same division can be applied to knives (Ballin 2012, 2017). While felsite could be used to produce other tools, the production of axeheads and knives—both prestige and utilitarian—was the primary purpose for its use.

Distribution of felsite artifacts at parish and site scales Three techniques were used to visualize the distribution of artifacts across the islands (Figure 4). Where more detailed provenances were available, graduated symbols were used to show locations where multiple tools were found, as assemblages, hoards, or individually (Figure 4(B)). Where only coarse parish scale mapping was possible, the distribution of artifacts across the archipelago by parish was carried out. To account for the differential size of parishes, the number of tools were normalized using the total area of each parish (m2) and visualized as a red, amber, green (RAG) map (Figure 4C). Where more precise find locations were available, artifact distributions were interpolated using a weighted kernel density estimation (wKDE) (Conolly and Lake 2006). The kernel radius was calculated using a Gaussian rule-of-thumb estimation (8100 m for tools and 7320 m for sites) (Silverman 1986). This tool created a smooth estimation of tool density across the archipelago, allowing us to better understand the wider landscape distribution of material (Figure 4D). While remaining conscious of the relatively coarse chronology of Neolithic Shetland, by layering known Neolithic sites over the distribution, it was possible to compare the distribution of felsite tools with likely contemporary activity.

Neolithic coastal change and sea levels Islands are unique and singular places, subject to change from many factors including climatic, environmental, and anthropogenic. The coast of Shetland has changed since the Neolithic (e.g., Moore and Wilson 2014), although very little work has been done on Holocene sea level history of the islands. Regional studies (Sturt, Garrow, and Bradley 2013) suggest quite rapid sea level change in the Mesolithic, leveling out in the Neolithic period (Lambeck 1993). Evidence for continued, albeit slower sea level rise, is visible in submerged peat deposits which can be found across the archipelago. This rapid decrease in sea level rise between the Mesolithic and Neolithic is visible in the islands’ paleoecological record. At harbor, on , radiocarbon dates from submerged peat indicate a coastline 8.9 m below present in the Mesolithic (Flinn 1964; Hoppe, Fries, and Quennerstedt 1965). Evidence from the Bridge of Walls near Scord of Brouster, West Mainland suggests that it had risen to 2 m below present levels by the Bronze Age (Bondevik et al. 2005). Given this leveling out it is likely that during the Neolithic the sea level was ca. 3–4 m lower than at present. This would mean that a great number of potential Neolithic coastal sites contemporary with the first quarrying activity at the felsite quarry complex are now underwater. Based on a 3–4 m sea level rise, bathymetric data from the UK Hydrographic Office suggests varying levels of inun- dation along the coastline. The main islands in the archipelago (including Yell, , Whalsay, and ) were still islands as larger water bodies like , Yell THE JOURNAL OF ISLAND AND COASTAL ARCHAEOLOGY 9

Figure 4. (A) Neolithic sites; (B) Graduated symbols representing number of felsite tools and their find locations (n ¼ 205); (C) Number of felsite tools per parish, normalized by km2 (n ¼ 267); (D) KDE of felsite tool distribution (n ¼ 205) and Neolithic sites (from Canmore). 10 W. P. MEGARRY ET AL.

Sound, and were sufficiently deep to remain open. Many smaller bays and inlets, including Brea Wick would likely have been lakes or brackish tidal zones (Bondevik et al. 2005). The extent of submersion of coastal areas is equally localized. The gentler bathymetric gradient along the southern coast of Eshaness, North Mainland suggests several hundred meters of submerged coastline, while the steeper northern coast of West Mainland indicates a shorter coastal zone. While the impact on the Neolithic coastal zone is variable and potentially extreme, the topography of the wider landscape would have been similar to that of the present day.

Modeling terrestrial and maritime movement Felsite was exclusively quarried from sites in North Roe, yet tools are found in consider- able numbers across the archipelago. As mentioned above, the cha^ıne operatoire of the quarrying and production process is now better understood following NRFP excavations in North Roe. Here, material was extracted and reduced and roughouts were produced on-site. Further stages appear to have taken place at sites away from the quarry. As dis- cussed above, it is very likely that most coastal sites are now underwater; however, there is some surviving evidence of stone working from sites at a remove from the Neolithic coast- line. Material from sites at Tangwick, Flugarth, Punds (NRFP analysis), and Crooksetter (Ballin 2017), all in North Mainland and Modesty (Kinghorn 1894), West Mainland sug- gests either primary or secondary grinding/polishing of felsite tools, indicating the trans- portation of material from the quarries to these sites. Understanding this transportation network can tell us much about the Neolithic Shetland landscape. Computational model- ing of journey times and paths was undertaken to model this dynamic (Figure 5). This included least-cost surfaces for journey times and paths from the quarry complex to the sites at Punds, Tangwick, and Modesty. Modeling movement computationally is not with- out issues (Bell and Lock 2000; Collischonn and Pilar 2000), and care needs to be taken to ensure an appropriate algorithm, sympathetic to the local landscape, is used. For terrestrial movement, a model based on anisotropic friction costs calculated for Scottish mountain- ous landscape was used and compared to actual journey times experienced during the NRFP (Aitken 1977; Langmuir 1984). In accordance with paleoecological records, which indicate a relatively open landscape on Shetland at this time (Bennett et al. 1992; Edwards 2014; Edwards and Whittington 1998), no additional costs were added for vegetation. While there has been considerable research into terrestrial movement in the land- scape, with some notable exceptions (Blankshein 2019; Callaghan and Scarre 2009; Garrow and Sturt 2011), maritime movement across and between islands is less well explored, especially in Northwest Europe. In Shetland, it seems likely that felsite was being transported off North Roe on foot (perhaps to sites like Flugarth), and then around the archipelago utilizing both terrestrial and short maritime journeys. While dis- tance, wind, and weather were important factors over longer distances, for shorter jour- neys the critical importance of local tides and maritime knowledge cannot be overemphasized (Blankshein 2019; Callaghan and Scarre 2009). Much would also have depended on the boats used. Little is known of Neolithic maritime watercraft, but it is generally accepted that these were skin or hide boats, propelled using paddles (McGrail 2004; Van de Noort 2011). This study assumes journeys across short stretches of water THE JOURNAL OF ISLAND AND COASTAL ARCHAEOLOGY 11

Figure 5. (A) Least-cost surface showing journey times; (B) Coastal exposure along the Northmavine coast (Flinn 1964); and (C) Paths from the Ronas quarries to felsite working sites mentioned in text. Surfaces generated using the r.walk anisotropic function (after Fontenari et al. 2005). and along coastlines, on small paddle-propelled hide boats using prevailing tides. These models are not intended to exactly replicate past actions or experiences. Rather, they are useful to explore the interaction between past peoples and their landscapes in more general terms and have been used effectively to look at issues ranging from hunter-gath- erer resource procurement over wide areas (Rissetto 2012) to territory demarcation on islands (Beavan 2011).

Results: Understanding the role of felsite in Neolithic Shetland Results from the project will be presented in this section which explores what the distribu- tion of felsite polished stone tools can tell us about wider Neolithic society in Shetland. It also explores what it can tell us about the production of tools away from the quarry sites.

Neolithic society in Shetland The results challenge existing models of Neolithic settlement in Shetland which have been centered on a small number of excavations at settlements on West Mainland. The distribu- tion of tools across the archipelago indicates a far more extensive use of the landscape. Tools were transported from quarry sites in Northmavine, North Mainland, to Sumburgh at the southern tip of Mainland, to Unst, the most northerly island in the archipelago. They also reach smaller islands like Whalsay. It is perhaps unsurprising that the greatest 12 W. P. MEGARRY ET AL. quantity (105 or 27%) of felsite tools have been found in Northmavine (Figure 4A, C). What is perhaps more surprising is that the densest concentrations of felsite tools, when the area of each parish is taken into account, are found in Aithsting in West Mainland and the island of Whalsay, some distance from the quarries at North Roe (Figure 4A, B). Figure 4D uses existing known Neolithic site locations as a comparison for exploring fel- site tool distribution. Comparing background site and tool datasets indicates some inter- esting distinctions. Clusters or concentrations of felsite tools exist in areas with relatively few known contemporary sites. Of most interest in this comparison is the concentration of tools around St. Magnus Bay, between Northmavine and West Mainland and, more specif- ically, the difference discernible within the area of West Mainland. While the site distribu- tion here is focused on central and southern areas, felsite tool distribution is focused toward the northern coast and the area of Modesty and West Burrafirth (Figures 1 and 4D). This pattern can be seen elsewhere in Shetland. While some find locations of tools correspond to known Neolithic sites, many do not. In these cases, the key implication is that the presence of felsite can be taken as a potential proxy for settlement activity during the period. This provides an important approach in targetting areas of Neolithic activity, potentially from early in the Neolithic.

Polished stone tool production on Shetland There is no evidence for either grinding or polishing at the quarries; however, there are strong indications that this was happening at sites away from the quarries. With a few exceptions, the locations of these sites have remained elusive, in part at least due to the changes in sea level discussed above. Field survey undertaken as part of the NRFP has gone some way to address this issue. This included a site at Flugarth, 4 km east of the quar- ries at the Beorgs of Uyea where 40 felsite objects including debitage, scrapers, a roughout and a small hammerstone were collected during fieldwalking in 2018. Analysis of this assemblage by Gilhooly (2019) suggest early, middle, and late stage roughing out, grinding and polishing, and activities indicative of a habitation or occupation site. The location of this site on the route from the Beorgs of Uyea to the eastern coast is of considerable sig- nificance. Similar evidence for polishing is visible from other sites identified by the project including a small felsite scatter at Tangwick, and from known assemblages at Punds (NRFP analysis), Modesty (Kinghorn 1894; Sheridan 2012), and Crooksetter (Ballin 2017). While Ballin (2017) interprets these sites as “chop shops” where recycling of felsite tools took place, some of the objects may represent primary production using roughouts from the quarry sites, further supporting the hypothesis that the cha^ıne operatoire of polished stone tool production is spread, at least across Mainland (Figure 6).

Felsite transportation networks on Shetland The distribution of felsite suggests a clear concentration of activity (Figure 6) in the coastal areas around St Magnus Bay. More broadly, there is a focus on the mid-Mainland belt, extending east to the island of Whalsay (Figure 4D). While coastlines may have changed by up to 4 m, the wider distribution of material in these areas suggest a focus of Neolithic activity. Decreasing numbers of tools are found on the islands of Yell and Unst THE JOURNAL OF ISLAND AND COASTAL ARCHAEOLOGY 13

Figure 6. Examples of tools from , , and Modesty, including roughouts from Eschaness and tool distributions around St Magnus Bay. to the north and toward the southern end of Mainland. In fact, very few artifacts have been found to date in coastal locations in North Roe, immediately west and north of the quarries. Here the shoreline is exposed and treacherous for boats (Figure 5). Indeed, much of the coastline in Northmavine is dangerous, and the small number of beaches east of North Roe and on southern Eshaness would have made attractive harbors. Based on models terrestrial travel times from the quarries to coastal sites at Punds, on Voe on the east side of Mainland and Tangwick on Eshaness suggest rela- tively short journeys of around four to five hours. Longer journeys, like the terrestrial route to Modesty would have been far more time-consuming, taking days (Figure 5). Given these constraints, it seems highly likely that material was brought terrestrially to locations like Flugarth on the east of North Roe or coastal areas on Eshaness and fur- ther east to or Punds. From Flugarth and Punds it must then have been transported by boat across to Yell and then further north to Unst, or from sites on the southern coast of Eshaness by boat along the coast around St Magnus Bay to settle- ments in West Mainland. The details are less clear on how tools were transported to the south of Mainland and other islands, but a similar combination of terrestrial and maritime methods must have been used.

Discussion: Felsite and neolithic society in Shetland Analysis is only as good as the data on which it is based, and geospatial analysis is no differ- ent. While significant care was taken to compile a robust and detailed catalogue of tools, the 14 W. P. MEGARRY ET AL. techniques applied in this paper are impacted by the biases inherent in any archaeological dataset including mixed scales of provenance. Given the aim of this paper to explore the role of felsite at an archipelago-scale, the techniques employed to address them focused on the wider landscape rather than individual sites. Results indicate that felsite tools were used across the archipelago, including places where no other contemporary evidence has been recognized. While sites are used as a proxy for known activity, changes in sea level has impacted on our understanding of the Neolithic coast. In spite of this, the felsite tool distri- bution can still act as indicator of activity. The distribution of tools (Figures 4, 6, and 7) indicate concentrations of activity around St Magnus Bay, specifically in southern Eschaness and along the northern coast of West Mainland. While there are known sites in this area, it seems likely that there is substantially more activity here than previously believed. Targeted research, including non- invasive aerial and geophysical prospection and excavation, in these areas has great potential to define the character and chronology of the use of felsite use and the Neolithic in general.

Tools and chronology As discussed in the introduction, excavations and radiocarbon dates from West Mainland have resulted in a strong regional and chronological bias. Recent dates (Table 1) suggest that felsite tool production was well established in the Early Neolithic (Cooney et al. 2019; Cummings 2017; Sheridan 2012). Felsite continues to be used throughout the Neolithic and into the Bronze Age (Ballin 2012; Downes and Lamb 2000; Whittle 1986). The traditional understanding of Neolithic in Shetland is focused on West Mainland, but the distribution of felsite suggests a more complex archaeo- logical landscape, with widespread activity across the archipelago. This appears to be further supported, at least in the later Neolithic, by the heel-shaped burial cairns which have been recorded across the islands (Henshall 1972; Mahler 2012; Sheridan 2012). Unfortunately, very few have been properly excavated and none—with the exception of two cairns at ’s Voe which are Early Bronze Age—have been securely dated (Sheridan 2012). What is clear is that felsite tool production was well established and operating on an impressive scale from early in the Neolithic (Cooney et al. 2019). This included the production of both axes and knives, which have been found at sites across Mainland and can be dated to this period (Ballin 2017; Cummings 2017; Kinghorn 1894; Sheridan 2012).

Quarries and the island landscape Neolithic quarries such as the North Roe quarry complex represent major investment of time and labor; however, they have not conventionally been seen as equivalent to the effort invested in monuments elsewhere in the Neolithic world. Quarries are often located in what are regarded as liminal, marginal places like islands or high in moun- tainous regions (Bradley et al. 1992; Cooney 2017; Cooney, Warren, and Ballin 2013; Petrequin et al. 2012). These places took time and effort to reach, yet they become the centers of local, regional and even international distribution networks. In Shetland, these networks are archipelago-wide while other quarries like those for Group VI THE JOURNAL OF ISLAND AND COASTAL ARCHAEOLOGY 15

Figure 7. Recorded Neolithic sites (Source: Canmore); kernel distribution of sites (7320 m kernel, n ¼ 186); Stone tool locations (Source: NRFP); kernel distribution of find locations with weighting based on number of tools (8100 m kernel, n ¼ 290). 16 W. P. MEGARRY ET AL. volcanic tuff axes from Cumbria or Group IX porcellanite axes from Tievebulliagh and Rathlin island in County Antrim, Northern Ireland are the center of regional networks across Ireland and Britain (Bradley and Edmonds 1993; Grogan and Cooney 1990). The multi-scalar nature of quarries and quarrying is what makes them so valuable as they allow us to simultaneously explore the dichotomy of islands as both bounded units and as components of complex wider maritime landscapes (Cooney 2017). The interpret- ation of stone knives and axes as ceremonial as well as utilitarian tools is not just based on the labor that went into their production and polish, but also they were special objects from special places. Archaeological evidence from the Beorgs of Uyea in North Roe suggest that the quarry landscape itself was highly ritualized. For example, a small burial cairn (Henshall 1972) was constructed at the site and over 50 small (<1 m) red granite and quarried felsite standing stones were erected on the summit above the quar- ries, demarcating the extent of exploitation to the east and south (Cooney et al. 2019; Megarry et al. 2018). This is a place that was linked to the expression of identity through the acquisition, working and distribution of felsite across the archipelago.

Movement within and outwith Shetland Like other Neolithic Scottish islands, the sea clearly played a central role in communica- tions in Neolithic Shetland (Garrow et al. 2017; Garrow and Sturt 2011; Van de Noort 2011). Models suggest that terrestrial travel would have been preferred to move material to suitably sheltered harbors, but would have been less suitable for longer distances where short distances over familiar stretches of water were preferred. This dynamic is particularly clear around St Magnus Bay and on the eastern coast of Northmavine where material was transported along the coast and across shorter stretches of open water for distribution across the archipelago. St Magnus Bay would also have provided opportunities for shelter as material was moved from the quarries in North Roe to com- munities in West Mainland and further south, and this was clearly quicker than longer terrestrial journeys. The presence of large numbers of felsite tools across Shetland, including more distant islands like Yell, Unst, and Whalsay, clearly indicates local mari- time transport networks. Based on the distribution of felsite tools, it is suggested that there were three main distribution routes on Neolithic Shetland (Figure 8). Along one route felsite roughouts and tools were transported across the landscape to Eshaness where they were moved by boat across (or around) St Magnus Bay to communities in West Mainland. A second route likely moved material across the relatively short stretch of water from East Mainland to Yell, while a third terrestrial route moved tools south- ward across central Mainland. One of the distinctive features of Neolithic Shetland is that felsite was clearly access- ible and used right across the archipelago. The work of the NRFP has established that the production, working and use of felsite for axes and knives was a feature of life from the establishment of the Neolithic around 3600 BC. The sparsity of felsite artifacts found outside Shetland (only two axes) is worthy of some discussion, as is the near total absence of externally produced axes or lithologies (a single exception to this is discussed below). Polished axes particular sources are widely distributed over large areas within the wider western European Neolithic, where they are often treated as prestige goods THE JOURNAL OF ISLAND AND COASTAL ARCHAEOLOGY 17

Figure 8. Terrestrial and maritime general routes for felsite on Neolithic Shetland with Neolithic sites as suggested by the study. See Figure 1 for key sites. 18 W. P. MEGARRY ET AL.

(Bradley and Edmonds 1993; Cooney 2015). Quarries are often found in marginal envi- ronments like islands or mountainsides yet physical geography was rarely a barrier to distribution. The location of the Shetland islands 160 km from mainland Scotland cer- tainly pushes this dynamic to the extreme, yet the islands were inter-visible with Orkney and the mainland via and clear evidence for contact bookends the period. A single porcellanite ax (ARC 65256: Shetland Museum) remains the only evi- dence for external contact from the NRFP database. Quarried at Tievebulliagh or Rathlin island, Northern Ireland (Cooney 2017), porcellanite is found along the Atlantic fac¸ade of Scotland. Other aspects of Neolithic life similarly suggest a degree of separation after an initial period of contact. Sheridan (2012) has noted the complete absence of regional pottery typologies on Shetland until the late Neolithic and Early Bronze Age. Throughout the Neolithic and into the Bronze Age, distinctive chambered cairns evolved to include an island-specific triad of features including a passage, a trefoil chamber, and a distinctive heeled fac¸ade which, according to Sheridan (2012), is often constructed during a later phase (Hamilton 1956; Henshall 1972). Notably absent from Shetland are long-cairn additions to monuments found elsewhere in Northern Scotland. Kaul (2011) has sug- gested that dating these additions may provide a terminus ante quem when Shetland diverted from the regional Neolithic. This divergence, following an initial period of con- tact, suggests a dynamic not entirely explained by geographical distance from Orkney. It is proposed here that the styles, patterns, and extent of the use of felsite suggest a “process of self-identification” (Robb 2001), where insular identity is intentionally mani- fest in distinctive material culture and practices, originally based on (but with quick adaptation from) preexisting regional styles. This is particularly visible in the Shetland knives which rapidly reach a scale and sophistication not seen amongst comparanda elsewhere (Clarke 2016). This process is likely similar to distinctive cultural expressions on other islands, like the Maltese temple culture, which had its origins in the subterra- nean hypogea found in southern Italy (Robb 2001), or the Moai and Ahu of Rapa Nui, which find their roots in the marae common across Polynesia (Hamilton, Thomas, and Whitehouse 2011; Hunt and Lipo 2011; Roger 2010). In all cases, insularity quite rapidly became a driver for innovation and self-identification.

Conclusion This article is part of a special issue revisiting the concept of the “island laboratory.” Recent decades have seen many challenges to traditional biogeographical interpretations of island cultures by exploring the relationship between material culture and identity, emphasizing the social experiences embedded within these communities. This paper fur- ther expands on these themes, proposing that Neolithic quarries, and stone tool produc- tion and distribution in particular, allow us to explore both the physical and social geographies of past island cultures. Often located in liminal or hard-to-reach locations, quarries are nevertheless centers of complex economic and social networks. The distri- bution of tools from quarries provide a window into Neolithic life in Shetland while their absence further afield allows us explore questions of mobility, identity and choice. The presence of felsite tools across the archipelago suggests a far more extensive and THE JOURNAL OF ISLAND AND COASTAL ARCHAEOLOGY 19 geographically distributed community than previously thought. As the primary lithology used, felsite was sourced from a specific quarry complex which became the hub of an island-wide distribution network, well-established by the middle of the fourth millen- nium BC. This network was part of the cha^ıne operatoire of felsite tool production with the initial stages of production at the quarry but the final stages at often distant loca- tions. It also shows the degree of interaction between the Neolithic population and the “islandscape” (Broodbank 2000), and the integration of physical and social dimensions within the archipelago “laboratory.” Models suggest both terrestrial and maritime modes of transportation with a preference for the latter, where the sea played a role as hyper- conductor, facilitating movement. Conversely, the sea may also have acted as an insula- tor, facilitating the creation of a distinct insular dynamic, particular to these islands.

Acknowledgements The authors would like to thank Helen Dawson, Llorenc¸ Picornell Gelabert, Manel Calvo, Biel Servera and Alex Valenzuela as guest editors of this special issue, and for organizing the session, “The ‘island laboratory’ revisited: Integrating environmental and socio-cultural approaches” at the European Association of Archaeologists (EAA) in 2018, where this paper was initially presented. We would also like to thank Prof Todd Braje, journal editor, and the two anonymous peer reviewers who provided such helpful feedback, and undoubtedly improved the manuscript. Thanks also to the North Roe Felsite Project (NRFP) team members over the years including Torben Ballin, David Field, Joanne Gaffney, Astrid Nyland, Jostein Gunderson, Niamh Kelly, Rob Sands, Alison Sheridan, Pete Topping, and the late Vin Davis. Support was always available from the staff at the Shetland Museum and Archives, especially Jenny Murray. Val Turner and Chris Dyer at the Shetland Amenity Trust also provided invaluable local assistance and advice. Thanks are also due to landowners who supported the project including Jane Brown, John Alec Cromarty, and Mona Walterson. We would also like to acknowledge and thank Archaeology Shetland for their contribution to the project and to Nick Card and Prof Vicki Cummings for allowing us to reproduce radiocarbon dates from their excavations at Crooksetter.

Disclosure statement The authors declare no conflict of interest.

Funding The authors also acknowledge the financial support of the National Geographic Global Exploration Fund (Northern Europe) Grant Numbers; GEFNE54-12; GEFNE 147-15. The Culture and Society Cluster at Queen’s University Belfast and UCD College of Social Sciences and Law and the UCD School of Archaeology at University College Dublin also provided finan- cial support.

References Aitken, R. 1977. Wilderness areas in Scotland. PhD diss., University of Aberdeen. Ballin, T. 2011. The felsite quarries of North Roe, Shetland: An overview. In Stone Axe studies III, ed. V. Davis and M. Edmonds, 121–30. Oxford: Oxford Books. 20 W. P. MEGARRY ET AL.

Ballin, T. B. 2012. Felsite axehead reduction: The flow from quarry pit to discard/deposition. In The border of farming Shetland and Scandinavia: Neolithic and Bronze Age farming, ed. D. Mahler and C. Anderson, 73–91. Copenhagen: The National Museum of Denmark. Ballin, T. B. 2017. North Roe felsite and Shetland’s felsite material culture - Assemblages from quarry workshops and domestic settlements. In Written in stone. Papers on the function, form, and provenancing of prehistoric stone objects in memory of Fiona Roe, ed. R. Shaffrey, 77–110. St Andrews: Highfield Press. Ballin, T. B., and D. Mahler. 2017. Ardpoint (sandstone) quarry in West Mainland (Shetland Is- lands; Walls and Sandness parish). Discovery and Excavation in Scotland. New Series 17 (2016): 161–2. Barcham, R. 1978. A lost radiocarbon date for Shetland. Proceedings of the Society of Antiquaries of Scotland 110:502–6. Bevan, A. 2011. Computational models for understanding movement and territory. In Tecnologıas de informacıon geografica y analisis arqueologico del territorio: Actas del V Simposio Internacional de ArqueologıadeMerida [Geographic information technologies and archaeological analysis of territories: Proceedings of the 5th International Symposium of Archaeology at Merida, Spain], ed V. Mayoral Herrera and S. Celestino Perez S, 383–94. Merida: Council of Scientific Investigations. Bell, T., and G. Lock. 2000. Topographic and cultural influences on walking the Ridgeway in later prehistoric times. In Beyond the map. Archaeology and spatial technologies, ed. G. Lock, 85–100. Amsterdam: IOS Press/Ohmsha. Bennett, K. D., S. Boreham, M. J. Sharp, and V. R. Switsur. 1992. Holocene history of environ- ment, vegetation and human settlement on Catta Ness, Lunnasting, Shetland. The Journal of Ecology 80 (2):241–73. doi:10.2307/2261010 Blankshein, S. 2019. Transient trajectories: Modelling movement and connectivity in the Neolithic of the Outer Hebrides. PhD diss., University of Southampton. Bondevik, S., J. Mangerud, Ø. Lohne, S. Dawson, and A. Dawson. 2005. Evidence for three tsunamis at the Shetland Islands between 8000 and 1500 years ago. Quaternary Science Reviews 24 (14-15):1757–75. doi:10.1016/j.quascirev.2004.10.018 Bradley, R., and M. Edmonds. 1993. Interpreting the axe trade: Production and exchange in Neolithic Britain. Cambridge: Cambridge University Press. Bradley, R., P. Meredith, J. Smith, and M. Edmonds. 1992. Rock physics and the Neolithic axe trade in Great Britain. Archaeometry 34 (2):223–33. doi:10.1111/j.1475-4754.1992.tb00494.x Broodbank, C. 2000. An Island Archaeology of the Early Cyclades. pp. 6–35. Cambridge: Cambridge University Press. Calder, C. S. T. 1949. Report on the excavation of a Neolithic temple at Stanydale in the parish of , Shetland. Proceedings of the Society of Antiquaries of Scotland 84:185–205. Calder, C. S. T. 1956. Report of the discovery of numerous Stone Age house-sites in Shetland. Proceedings of the Society of Antiquities of Scotland 89:340–97. Callaghan, R., and C. Scarre. 2009. Simulating the Western Seaways. Oxford Journal of Archaeology 28 (4):357–72. doi:10.1111/j.1468-0092.2009.00333.x Clarke, A. 2016. The Skaill knife as a butchering tool. Lithics - The Journal of the Lithic Studies Society 10:16–27. Collischonn, W., and J. V. Pilar. 2000. A direction dependent least-cost-path algorithm for roads and canals. International Journal of Geographical Information Science 14 (4):397–406. doi:10. 1080/13658810050024304 Conolly, J., and M. Lake. 2006. Geographical information systems in archaeology. Cambridge: Cambridge University Press. Cooney, G. 2015. Stone and flint axes in Neolithic Europe. In The Oxford handbook of Neolithic Europe, ed. C. Fowler, J. Harding, and D. Hofmann, 515–34. Oxford: Oxforn University Press. Cooney, G. 2017. The role of stone in Island societies in Neolithic Atlantic Europe: Creating pla- ces and cultural landscapes. Arctic 69 (5):1–12. doi:10.14430/arctic4666 THE JOURNAL OF ISLAND AND COASTAL ARCHAEOLOGY 21

Cooney, G., T. Ballin, and W. Megarry. 2012. Objects for an island world, North Roe Felsite, Shetland. In The border of farming and the cultural markers, ed. D. L. Mahler, 87–93. Copenhagen: The National Museum of Denmark. Cooney, G., W. Megarry, M. Markham, B. Gilhooly, B. O’Neill, J. Gaffrey, R. Sands, A. Nyland, T. B. Ballin, J. Murray, et al. 2019. Tangled up in blue: The role of riebeckite felsite in Neolithic Shetland. In Mining and quarrying in Neolithic Europe: A social perspective, ed. A. Teather, P. Topping, and J. Baczkowski, 49–65. Oxford: Oxbow Books. Cooney, G., G. Warren, and T. Ballin. 2013. Island quarries, island axeheads, and the Neolithic of Ireland and Britain. North American Archaeologist 34 (4):409–31. doi:10.2190/NA.34.4.g Cummings, V. 2017. The Neolithic of Britain and Ireland. Oxford: Routledge. Downes, J., and R. Lamb. 2000. Prehistoric houses at Sumburgh in Shetland: Excavations at Sumburgh airport 1967-74. Oxford: Oxbow Books. Edmonds, M. 2019. Orcadia: Land, sea and stone in Neolithic Orkney. London: Head of Zeus Ltd. Edwards, K. J. 2014. Early farming, pollen and landscape impacts from northern Europe to the North Atlantic: Conundrums. In Northern worlds - Landscapes, interactions and dynamics, ed. H. C. Gulløv, 189–201. Copenhagen: National Museum of Denmark. Edwards, K. J., and G. Whittington. 1998. Landscape and environment in prehistoric West Mainland, Shetland. Landscape History 20 (1):5–17. doi:10.1080/01433768.1998.10594498 Flinn, D. 1964. Coastal and submarine features around the Shetland Islands. Proceedings of the Geologists’ Association 75 (3):321–39. doi:10.1016/S0016-7878(64)80036-5 Fontenari, S., Franceschetti, D. Sorrentino, F. Mussi, M. Pasolli, M. Napolitano, and R. Flor. 2005. r.walk. GRASS GIS manual page. https://grass.osgeo.org/grass70/manuals/r.walk.html. Garrow, D., S. Griffiths, H. Anderson-Whymark, and F. Sturt. 2017. Stepping stones to the Neolithic? Radiocarbon dating the Early Neolithic on islands within the “Western Seaways” of Britain. Proceedings of the Prehistoric Society 83:97–135. doi:10.1017/ppr.2017.4 Garrow, D., and F. Sturt. 2011. Grey waters bright with Neolithic argonauts? Maritime connec- tions and the Mesolithic-Neolithic transition within the “Western Seaways” of Britain, c. 5000–3500 BC. Antiquity 85 (327):59–72. doi:10.1017/S0003598X00067430 Gilhooly, B. 2018. An axe for all seasons; An investigation into the construction methods, mech- anical properties and uses of Irish shale and porcellanite axes from the Mesolithic to the Bronze Age, through experimental, quantitative, and comparative studies. PhD diss., University College Dublin. Gilhooly, B. 2019. Analysis of felsite lithic assemblage from Flugarth, North Roe, 2018 field sea- son. Unpublished field report. University College Dublin, Dublin. Gillmore, G. K., and N. Melton. 2011. Early Neolithic sands at West Voe, Shetland Islands: Implications for human settlement. Geological Society, London, Special Publications 352 (1): 69–83. doi:10.1144/SP352.6 Grogan, E., and G. Cooney. 1990. A preliminary distribution map of stone axes in Ireland. Antiquity 64 (244):559–61. doi:10.1017/S0003598X00078443 Hamilton, J. R. C. 1956. Excavations at , Shetland. Ministry of Works Archaeological Reports No 1. Ministry of Works, Edinburgh. Hamilton, S., M. Thomas, and R. Whitehouse. 2011. Say it with stone: Constructing with stones on Easter Island. World Archaeology 43 (2):167–90. doi:10.1080/00438243.2011.586273 Henshall, A. S. 1972. The chambered tombs of Scotland, 1 and 2. Edinburgh: Edinburgh University Press. Hoppe, G., M. Fries, and N. Quennerstedt. 1965. Submarine peat in the Shetland Islands. Geografiska Annaler: Series A, Physical Geography 47 (4):195–203. doi:10.1080/04353676.1965. 11879719 Hunt, T., and C. Lipo. 2011. The statues that walked: Unravelling the mystery of Easter Island. New York: Free Press. Kaul, F. 2011. Multi-period construction of megalithic tombs–and the megalithic tombs in Shetland. In The border of farming and the cultural markers, ed. D. L. Mahler, 100–21. Copenhagen: The National Museum of Denmark. 22 W. P. MEGARRY ET AL.

Kinghorn, G. 1894. Notes on the discovery of a deposit of polished stone axes and oval knives of porphyry at modesty, near Bridge of Walls, Shetland. Proceedings of the Society of Antiquaries of Scotland 29:48–54. Lambeck, K. 1993. Glacial rebound of the British Isles—I. Preliminary model results. Geophysical Journal International 115 (3):941–59. doi:10.1111/j.1365-246X.1993.tb01503.x Langmuir, E. 1984. Mountaincraft and leadership: A handbook for mountaineers and hillwalking leaders in the British Isles. Edinburgh: Scottish Sports Council. Mahler, D. L. D. 2012. Pinhoulland - A multi period site from West Mainland, Shetland. In The border of farming and the cultural markers, ed. D. L. Mahler, 37–52. Copenhagen: The National Museum of Denmark. Markham, M. 2018. Northmavine felsite and Shetland Neolithic tools: Geology and petrology. Unpublished report for the North Roe Felsite Project ‘Making and Island Work: Neolithic Shetland’. University College Dublin, Dublin. McGrail, S. 2004. Boats of the world: From the Stone Age to medieval times. Oxford: Oxford University Press. Megarry, W., C. Graham, B. Gilhooly, B. O’Neill, R. Sands, A. Nyland, and G. Cooney. 2018. Debitage and drones: Classifying and characterising Neolithic stone tool production in the Shetland Islands using high resolution unmanned aerial vehicle imagery. Drones 2 (2):12. doi: 10.3390/drones2020012 Megarry, W. P., G. Cooney, D. C. Comer, and C. E. Priebe. 2016. Posterior probability modeling and image classification for archaeological site prospection: Building a survey efficacy model for identifying Neolithic felsite workshops in the Shetland Islands. Remote Sensing 8 (6):529. doi:10.3390/rs8060529 Melton, N. D. 2008. West Voe: A Mesolithic–Neolithic transition site in Shetland. In Scottish odysseys: The archaeology of islands, ed. G. Noble, T. Poller, J. Raven and L. Verrill, 23–36. Stroud: Tempus. Moore, H., and G. Wilson. 2014. Ebbing shores: Survey and excavation of coastal archaeology in Shetland 1995-2008. Edinburgh: Historic Scotland. Montgomery, J., J. Beaumont, M. Jay, K. Keefe, A. R. Gledhill, G. T. Cook, S. Dockrill, and N. Melton. 2013. Strategic and sporadic marine consumption at the onset of the Neolithic: Increasing temporal resolution in the isotope evidence. Antiquity 87 (338):1060–72. doi:10. 1017/S0003598X00049863 O’Neill, B., B. Gilhooly, and G. Cooney. 2014. Stone tools of Shetland: Experimental felsite pro- ject. EXARC Online Journal 3. https://exarc.net/ark:/88735/10163 Petrequin, P., S. Cassen, M. Errera, L. Klassen, A. Sheridan, and A. Petrequin, eds. 2012. Jade. Grandes haches alpines du Neolithique europeen. Ve et IVe millenaires av. J.-C [Jade. Large Alpine axeheads in Neolithic Europe. 5th-4th Millennia BC.] (Les Cahiers de la MSHE Ledoux 17, Serie Dynamiques Territoriales 6). Besanc¸on: Presses Universitaires de Franche-Comte and Centre de Recherche archeologique de la Vallee de l’Ain. Rissetto, J. D. 2012. Using least cost path analysis to reinterpret Late Upper Paleolithic hunter- gatherer procurement zones in Northern Spain. In Least cost analysis of social landscapes. Archaeological case studies, ed. D. White and S. Surface-Evans, 11–31. Salt Lake City: University of Utah Press. Ritchie, P. R., and J. G. Scott. 1988. The petrological identification of stone axes from Scotland. In Stone axe studies volume 2: The petrology of prehistoric stone implements from the British Isles, ed. T. H. McK Clough and W. A. Cummins, 85–97. London: Council for British Archaeology. Robb, J. 2001. Island identities: Ritual, travel and the creation of difference in Neolithic Malta. European Journal of Archaeology 4 (2):175–202. Roger, F. S. 2010. Island at the end of the world. The turbulent history of Easter Island. London: Faber and Faber Ltd. Sheridan, A. 2012. Neolithic Shetland: A view from the “mainland”.InThe border of farming and the cultural markers, ed. D. L. Mahler, 6–36. Copenhagen: The National Museum of Denmark. THE JOURNAL OF ISLAND AND COASTAL ARCHAEOLOGY 23

Sheridan, A. 2013. Plus c¸a change … ? Developments in Shetland, c 2500–1800 BC. In The border of farming. Shetland and Scandinavia: Neolithic and Bronze Age farming, ed. D. L. Mahler, 47–73. Copenhagen: The National Museum of Denmark. Silverman, B. W. 1986. Density estimation for statistics and data analysis. New York: Routledge. Sturt, F., D. Garrow, and S. Bradley. 2013. New models of North West European Holocene palaeogeography and inundation. Journal of Archaeological Science 40 (11):3963–76. doi:10. 1016/j.jas.2013.05.023 Turner, V., F. H. Chrystall, I. Simpson, and E. B. Guttmann. 2004. Form and function in Shetland prehistoric field systems. In Atlantic connections and adaptations: Economies, environ- ments and subsistence in lands bordering the North Atlantic. Symposia of the Association for Environmental Archaeology, ed. R. Housley and G. Coles, vol. 21, 120–7. Oxford: Oxbow Books. Turner, V. E. 2012. Location, form and function in Shetland’s prehistoric field systems. PhD diss., University of Stirling. Van de Noort, R. 2011. North Sea archaeologies: A maritime biography 10,000 BC-AD 1500. Oxford: Oxford University Press. Whittle, A. 1986. Scord of Brouster: An early agricultural settlement on Shetland: Excavations, 1977-1979, Oxford University Committee for Archaeology Monograph No. 9. Oxford: Oxford University School of Archaeology.