Resources, Production, and Trade in the Norse Shetland Juha Marttila* Abstract - In this paper, I use the research carried out by scholars in Shetland and the North Atlantic to analyze and discuss the resources available for the Norse in Shetland, including steatite, fishing, whales, seals, birds, animal husbandry, and cultivation of crops. I present the utilization and control of these resources in the context of trade in the North Atlantic, and also investigate the economic changes occurring during the late 1st and early 2nd millennium AD and their effects on the settlement in Shetland. *Department of Archaeology, University of Glasgow, Glasgow, Scotland, UK. Current address - 50 St. Abbs Way, Chapelhall, Aridrie, ML6 8WG Scotland, UK; juha.marttila@gmail.com. Introduction As more and more research on the economic development in northern Europe has been carried out (e.g., Gaimster 1992, Ingimundarsson 1992, Miller 1990), it has become increasingly evident that market and non-market trade coexisted in both the Viking Age and the Medieval period. In the non-market economy, goods and services were mainly bartered. The market economy consisted of organized trade of surplus for financial gain. A distinction can be made between inter-regional trade of low-bulk, high-value luxury goods and low-value, high-bulk necessities (staples). The increase in the long-range trade of the staple goods is important to an understanding of socioeconomic transformations and may define the Viking Age/Medieval period transition in the Norse North Atlantic (Fig. 1; Barrett et al. 2000:15). One of these socioeconomic developments was the appearance of a new artisan class. These combmakers, shoemakers, and metalworkers produced items which ordinary people could afford and use in their everyday lives (Hansen 2015:28). While this trade in staples (henceforth “the trade”) is well referenced in the Post-Medieval period, no references exist from the Viking Age (ca. 800–1050 AD), and only a few survive from the Late Norse 2016 Journal of the North Atlantic No. 29:1–20 Figure 1. A map of the North Atlantic. 2016 Journal of the North Atlantic No. 29 J. Marttila 2 period (ca. 1050–1450 AD) (Forster 2004:117). In the Saga of King Sverri of Norway (Sverrisaga), Shetlandic merchants are mentioned alongside the ones from Orkney, Faroes Isles, and Iceland as bringing goods that “… make this land richer, and we cannot do without” (Sephton 1994:129). The text is dated to 1186 and is, according to Smith (1984:7), the earliest reference to Shetlandic merchants. It occurred at a time when the trade was being concentrated and markets were being created in places such as Lübeck and Bergen (Forte et al. 2005:174). The Hanseatic League became an increasingly dominant force in northern Europe during the 13th century, and its power increased as the control of the Norwegian Crown declined. The first Norwegian treaty with Lübeck was signed in 1250 (Smith 1984:7). Shetland is mentioned in only a few references in the next few centuries. In 1316, a Shetlandic burgess is known to have been in Bergen, whose presence can be seen as an indication of Shetland’s involvement in the trade routes focused on Bergen. This Shetlandic–Norwegian trade seems to have been concentrated on basic raw materials and commodities whose production entailed relatively little labor, such as stockfish (i.e., air-dried cod), timber, and cereals (Helle 1995:47, Lopez 1976:95, Smith 1984:7). By the 16th century, Shetland had been incorporated into the large, commercial European markets and traded stockfish, other fish products, and woollens with the Hanseatic merchants (Bigelow 1992:18). However, due to the lack of historical sources and difficulties in differentiating Shetlandic and Norwegian products, it is problematic to identify when the long-range trade of true staples first occurred (Barrett et al. 2000:16). It is possible that the staples, such as steatite blocks and vessels, and ragstone hones, were first transported as ballast, and the main highvalue commodity would have been, for example, imported walrus ivory (ibid.:16, Buckland and Sadler 1990:120, Dugmore et al. 2007). Even though historical records are very limited from the Viking and Late Norse periods, the material record can provide evidence of large-scale marketlevel interaction. Evidence such as resource control, the production of uniform commodities, and regular/ repetitive exchange can indicate commercial activities that occurred during these periods. Steatite Steatite is a soft metamorphic rock characterized by high hydrous magnesium silicate (talc) content. Due to its high talc content, it is easy to carve with metal or stone tools, and has a high resistance to heat (Buttler 1989:193, Turner 1998:95–97). The relatively easy working characteristics of steatite have been known by the people living in Shetland since the Neolithic period. Although it is not common to find a significant amount of steatite artifacts from prehistoric contexts, some artifacts are known (Buttler 1989:194, Turner 1998:95). For example, 4 small square-sided vessels were found from Jarlshof’s Bronze Age contexts (Hamilton 1956:20), and some from Kebister’s Iron Age levels (Owen and Lowe 1999:169). Steatite vessels are also known to have been used in funerary functions in the Northern Isles (Turner 1998:95). So far, only 1 prehistoric site has provided an extensive assemblage: recent excavations at Bayenne, Yell, provided a significant amount of vessel fragments dating from the Bronze and Early Iron Ages (Forster and Bond 2004:220). Besides the manufacture of vessels, steatite was also used as temper for pottery during the prehistoric periods, and pots with steatite inclusions are found all over Shetland (Buttler 1989:194, Turner 1998:95). The evidence suggests that, around the Middle Iron Age, the use of large steatite vessels declined, although the material was still used as a temper for pottery, and for lamps and portable objects (Forster and Bond 2004:220). Undoubtedly, the zenith of the use of steatite in the North Atlantic was the Viking Age. At the time, steatite vessels were Norway’s main domestic ware to the extent that the Norwegian material culture at the time has been classified as aceramic. As a result of the expansion of the Scandinavian culture with the migrating settlers, steatite artifacts were transported extensively (Bray et al. 2009:4). A steatite spindle whorl has been recovered even in North America at L’Anse aux Meadows, Newfoundland (Ingstad and Ingstad 2001:156). Sources of steatite outside Scandinavia in Shetland and Greenland were also utilized (see Figs. 1, 2). During the Viking Age, the production and distribution of steatite vessels in Scandinavia developed from modest beginnings into widespread distribution networks through which large quantities of affordable commodities were traded by the 10th century AD (Schou 2015:204). Steatite sources Scandinavia and Greenland. The majority of steatite sources in Scandinavia are located in Norway, although 7 quarries and other exposures are situated in Sweden and also some in Finland (Ritchie 1984:60, 64). Steatite quarries are abundant in Norway, and their distribution is widespread. Only Vestfold and Vest-Agder districts are marked by their absence (Ritchie 1984:61, Skjølsvold 1961:149 cited in Forster and Bond 2004:219). A considerable Journal of the North Atlantic 3 2016 No. 29 J. Marttila Figure 2. A map of northern Shetland. 2016 Journal of the North Atlantic No. 29 J. Marttila 4 concentration has been identified in Hardanger district, which appears to have been particularly important, especially in the Middle Ages (Sognnes 1979:223). The quarries consist of both large-scale excavations (see Skjølsvold 1969) and small domestic sites (Ritchie 1984:62–63). Steatite outcrops are abundant in Greenland, and the quality of the stone is exceptional (Bray et al. 2009:5). Despite the high quality of the stone, the Greenlandic steatite artifacts do not seem to have been not exported outside the source zone (Forster 2004:334). Shetland. Several metamorphic environments are situated in what is now Shetland resulted in the development of steatite. The majority of the outcrops are located on Unst (Fig. 3), but some also occur in Fetlar, Northmavine, Lunning (East Mainland), and South Mainland. Two Norse quarries have been excavated: Clibberswick in Unst (Buttler 1983) and Catpund in Cunningburgh (Turner 1998) (Forster and Bond 2004:220). Clibberswick has provided significant amount of evidence of Norse steatite working. The outcrop is located in northeast Shetland and includes 2 inlets, of which the more northerly of the 2 has been more intensively quarried. The north face of the promontory which separates the inlets (Cross Ness) is completely covered with bosses (Fig. 3) resulting from the removal of vessels (Bray et al. 2009:7). As the quarrying occurred high up on the rock faces and also well out on the sea cliffs, Ritchie (1981:11, 1984:63) suggests that they could have only been worked from a bosun’s chair slung from the cliff top or with a similar arrangement. A small structure has also been identified on top of the cliff, which provided evidence of the final-stage manufacture of the vessels (Buttler 1983:21–22). The place-name also indicates the presence of steatite, as it derives from the Old Norse kléa, to fasten stones (for example fishing weights) (Jakobsen 1928:vol. 1:427) or kleber, which is the Old Norse word for steatite. A substantial quarry area is located at Clammel Knowes, Unst, where a spoil heap including fragments of steatite has been identified on the downslope side. A limited number of bosses can be seen in Gorsendi Geo and Houllans Ness (Fig. 3). It seems as if small objects were worked at both these sites as well as in the other outcrops on the west coast of Unst including Wick of Collaster, Clay Geo Ness of Collaster, and Fiska Wick (Bray et al. 2009:7). Several sources of steatite at Fetlar were worked during the Viking Age and Late Norse Period (Fig. 4). Several circular or oval bosses, which represent the Norse manufacture of vessels, can be seen at Scarpi Geo, Hesta Ness. It is likely that the original quarrying was more extensive, but evidence has been destroyed due to quarrying after the First World Figure 3. Steatite sources in Unst. Figure 4. Steatite sources in Fetlar. Journal of the North Atlantic 5 2016 No. 29 J. Marttila War (ibid.:7, Ritchie 1984:73). On the south coast of Fetlar, the faces of Clemmil Geo and Dammins near Houbie were also extensively worked during the Norse settlement. Bosses of all shapes and sizes can be seen at both of the sites, and several spoil heaps have also been identified between them (Bray et al. 2009:7). The most extensive steatite workings at Northmavine are located in Fethaland on the Isle of Fethaland. Vessel bosses can be seen scattered around the outcrops on the steep, partially grass-covered slopes at Cleber Geo (Figs. 2, 5, and 6). A step appears to have been cut into the hillside at the steepest part of the area, possibly to provide a safer area for finishing the vessels. A significant amount of spoil is also visible at the quarry. The reworking of the landscape to such a scale is clear evidence of the intensity of the activities at the quarry (ibid.:7–8). Breibister, a short distance to the south of Cleber Geo, has also been extensively worked. Further south, the 2 outcrops at the Head of Calsta have provided evidence of limited workings. Cleber Geo at Hillswick Ness, in close vicinity to Niddister, has signs of steatite vessel production too. The working at the face must have been limited by the difficult access to the outcrop. It is possible, however, to pick up pieces of pink steatite from the beach at the eastern side of the Ness. At Orra Wick, Lunnin, which is located further south on the eastern side of Mainland Shetland, a number of bosses as well as a substantial vessel (which has not been detached) have been identified (ibid.:8–9). Catpund, Cunningsburgh, is located in the South Mainland of Shetland. It has been subjected to the most in-depth research of any steatite quarries in Shetland, and has proved to be the most extensively worked outcrop in the archipelago. The site consists of a large outcrop of talc-magnesite-schist, associated with serpentine and metamorphosed basic igneous rocks. Within this outcrop, there are several areas of good quality talc-magnesite rock, one of the largest and best-exposed being within the burn of Catpund, which covers an area of ~550,000 m2. In these deposits, the rock seems to be almost homogenous, and talc and magnesite are present in approximately equal amounts (Mykura 1976:119–120). This area has been known as a source of steatite for an extensive period of time, and hollows of Figure 5. Evidence of steatite workings at Cleber Geo from east . 2016 Journal of the North Atlantic No. 29 J. Marttila 6 either square or round shapes are the clearly visible results of quarrying all along Catpund burn (Turner 1998:95, 99; Turner et al. 2009). Considerable amounts of steatite were recovered during the excavations at Jarlshof, but Hamilton did not identify any manufacturing refuse at the site and therefore suggested that the vessels must have been imported from elsewhere (Hamilton 1956:206). However, Hamilton did not remove all deposits from the site, and manufacturing evidence has been recently identified (Colleen Batey, Department of Archaeology, University of Glasgow, Glasgow, Scotland, UK, pers. comm.). In any case, Catpund is the closest steatite source to Jarlshof and, as a result, Hamilton excavated the area in 1951. He recognized over 15 large mounds of spoil heaps of quarrying debris, of which he excavated 2. Using the small-find evidence, Hamilton (1956:206) suggested that 3 main vessel classes were manufactured in the area. In the 1980s, a detailed topographic survey was carried out in the area and was followed by a series of small excavations, which continued until 1990 (Turner 1998). There is evidence of quarrying from the natural outcrops and the areas adjacent to the burn. The burn also flows over some of the worked areas. There is clear evidence that the burn’s route had been altered in order to expose new areas for quarrying (Ibid.:99). There are significant spoil heaps to the south of the burn, which are located around flat, oval-shaped areas. It is possible that these areas could have been shafts sunk into the hillside in order for the steatite to be mined in the areas where the rock was situated deeper underground. Spoil heaps have been identified downslope of other Catpund hollows as well (Ritchie 1984:68–69, Turner 1998:99). If they represent deep mine shafts, however, it is unlikely that they would be filled to the extent as they appear today. In addition, some flat platforms appear to have been dug into the rock. This type of mining was also carried out at Østre Myre, Vegarshei, Norway (Buttler 1989:200). In 1988, a spoil tip of the quarry was excavated with unexpected results: the heap was excavated to bedrock and the bedrock showed evidence of quarrying as well. The area revealed by the excavation (8 m by 10 m) was densely covered in scars resulting from the removal of blocks for vessels. This suggests that Figure 6. A large boss at Cleber Geo. Journal of the North Atlantic 7 2016 No. 29 J. Marttila steatite was quarried all over the hillside and not just restricted in the areas where the rocks were visible. It seems, therefore, that the steatite had been worked in far larger scale than previously thought. This largescale working also explains the significant amount of debris on the hill. In some areas, the quarry had been worked in steps of up to 2 m in height (Turner 1998:99–100). It has been suggested that the steatite working at Catpund could have been of commercial nature (Ritchie 1984:70). The area exposed by the excavation contained over 160 holes within 5 distinctive working areas (which might represent separate quarrying events; Turner 1998:100): 1. Northern group - vessel blocks 2. West group - square vessel blocks 3. Southwest group - square vessel blocks 4. South group - rounded and/or rectangular blocks 5. East group - plates and square vessel blocks The holes varied from approximately 0.2 m square to 0.2 m x 0.7 m (oval). The division of the quarry into separate areas might be an indication of some degree of organization (i.e., commercial and/or seigniorial) of the industry (ibid.:100). The excavated spoil heap included waste from the quarrying process, half-finished broken vessels and plates, vessel and plate blanks (i.e., shaped externally but not hollowed out), and blocks (i.e., rock removed from the quarry floor which has not been shaped yet). The only tools found were a smoothing stone and some sandstone whetstones. As no small finds were discovered, it is likely that artifacts, such as loomweights and spindle whorls, were manufactured away from the site. Neither were any finished vessels found, which suggests that the vessels were hollowed out on the site in order to reduce the weight, and finished elsewhere. No finishing areas have been identified in close vicinity of the quarry so far (ibid.:100–101). The artifacts from the quarry varied greatly in the methods of tooling and in the shapes and thickness of the vessels, which would suggest that they were not worked by a skilled workforce, nor that the working of steatite was organized to any great extent. It is possible, however, that different individuals had their own areas of the quarry, as some divisions were revealed by the 1988 excavation (ibid.:101). This finding is possibly an indication of control of the resource and that permission was given to individuals to work certain areas of the outcrop. In 1990, several trenches were opened in the north area of the quarry, including a trench over a depression 3 m in diameter. The excavation revealed a 1.5-m-deep pit, which was stepped to the north, east, and south. The technique used in the removal of steatite was different in this area: instead of removing the steatite in vessel blocks, it seems to have been removed in sub-rectangular blocks of ~1 m by 0.5 m by 0.5 m. This method of working has also been found from Brennepösen, Akerhus (see Skjølsvold 1961) (Turner 1998:101–102). Recent research by Richard Jones of the University of Glasgow was able to differentiate the steatite from 3 different Norse quarries in Shetland. As Jones’ rare earth element analysis of steatite by ICPMS appears to be capable of separating sources in contrasting geological environments, the technique should provide crucial information as to the origin of steatite in the North Atlantic and facilitate the characterization of material culture deposits in Viking Age and Late Norse contexts (Jones et al. 2007:640). The typology of steatite vessels Norway. Amanda Forster has created a typology of Norwegian steatite vessels using the evidence from Norwegian sites such as Borgund, Sunnmøre (see Lossius 1977), Kaupang (see Forster 2004), and Gamlebyen, Oslo (see Lossius 1979). She also included the detailed study of ~3100 vessel sherds from Hedeby, northern Germany (see Resi 1979), as most of those artifacts originate from Norway (Forster 2004:147–149). The evidence from Norway indicates that the majority of the vessels from the Viking Age to the end of the 14th century are circular with a round base. Later, round vessels with a flat base were also manufactured (ibid.:157). These vessels were produced both in large scale for trade and in small scale for domestic use (ibid.:151). Other types of artifacts manufactured from steatite include baking plates, spindle whorls, weights, moulds, lamps, and tuyéres. These artifacts often are of secondary manufacture, made, for example, from broken vessel sherds. Consequently, their origins are difficult to determine and they will not be discussed here in more detail (ibid.:158–160, 162–163). Shetland. Numerous different types of vessels have been recovered in Shetland. Assemblages have been recovered from Jarlshof (Hamilton 1956), Old Scatness Broch (Forster and Bond 2004) Sandwick (Bigelow 1985:105–107), Underhoull (Small 1966:243), the Biggings, Papa Stour (Crawford and Ballin Smith 1999:131–133), and Norwick (Forster 2007:20). Forster (2004) has created a typology of Shetlandic vessels and bakeplates which consists of 5 main types. The forms of the vessels found in Shetland are similar to some extent to the ones found in Norway. In both areas, the most common form 2016 Journal of the North Atlantic No. 29 J. Marttila 8 semblage from this period consists of a high number of Norwegian imports, which are circular, hemispherical, and well-manufactured (ibid.:319–321). The Norse phase is characterized by the Norse adaptation to their local environment. The use of steatite varied depending on the access to the sources. The extensive importation from Norway declined together with the decline of the movement of the people, and it seems that the internal trade, which had developed within Scandinavia during the Viking Age, did not extend to the North Atlantic colonies (ibid.:323–324). In the areas without local sources of steatite, such as the Western Isles and the Faroe Isles, the diminished numbers of the vessels provided a stimulus for the development of locally produced ceramics (for the Faroese pottery see Arge 2000; Stummann Hansen 1991:51). In Iceland, steatite was partly replaced by the use of local stone, imported pottery, and wooden artifacts. In Shetland, the local resources were increasingly utilized, and Shetlandic vessel types developed. Steatite was not replaced in Orkney, but artifacts were imported from Shetland. For example, 4-sided steatite vessels have been recovered from Pool from contexts dated to the mid-10th and 11th centuries. The uniformity of the vessel type, together with its wide distribution, suggests a level of organization and might be an indication of the development of the steatite industry in Shetland. The bulk of the artifacts in Greenland were produced locally (Forster 2004:324–325). The Medieval phase is characterized by the continuous displacement of steatite in the areas without an easy access to a source and by the continuing importance of local resources where available. Steatite was still in use throughout the North Atlantic, although only in Shetland, Orkney, and Greenland had the use not diminished significantly since the Landnám phase. A limited number of artifacts were imported to the Faroe Isles and the Western Isles both from Norway and Shetland. The numbers from Shetland are not extensive, and the vessels seem to have originated from the Catpund quarry. The exportation of vessels and bakerstones was significantly more extensive to Orkney. The production at Catpund probably increased during this period, as the products were exported. The exportation suggests some control of the quarry and the production and distribution of its products. Within Shetland, there is evidence of the use of several quarries. The production probably declined by the end of the 14th century in response to increasing imports of ceramics, and thereafter production probably continued only for a smaller, domestic market (ibid.:327–330, 359). Imports from Norway have been recovered both from Orkney and Shetland. In Iceland, the finds is an undecorated, round-bottomed bowl (Buttler 1989:199). The most significant difference is the appearance of sub-rectangular (oval) and square vessels in Shetland, which seems to have been an indigenous development. As thick fragments of these vessels have been recovered from the Early Viking Age deposits from Old Scatness, this form seems to have been developed already by the 9th century. Four-sided vessels were recovered from prehistoric contexts from Bayenne as well, which might suggest that the structure of the stone dictated the shape of the vessels. For example, the majority of vessels from Old Scatness were manufactured from fine-grained steatite, but most of the sub-rectangular vessels were of significantly coarser stone. However, the development might have been functional as well. This type of vessel was more functional for long hearths which were common at the time. Thin-walled square vessels were introduced approximately at the end of the 11th century (Forster and Bond 2004:225–226). There seems to have been a general decline in the use of steatite during the Late Norse and Medieval periods in Shetland, as pottery became more common. For example, the Biggings, Papa Stour (Crawford and Ballin Smith 1999:145–146), Jarlshof (Hamilton 1956:Plate 85) and Sandwick (Bigelow 1985:107) have provided evidence of an increase in the use of pottery over the period of their inhabitation. The importation of continental redwares in particular seem to have had a great effect (Forster and Bond 2004:226). Other steatite artifacts found from Shetland include loomweights, line sinkers, spindle whorls, drill whorls, metalworking apparatuses, as well as trinkets and toys (Buttler 1989:195–198). The trade of steatite in the North Atlantic In order to understand the trade and origins of the steatite artifacts in the North Atlantic, Forster (2004) examined the assemblages recovered from the Faroe Isles, Iceland, Orkney, Caithness, the Western Isles, and the towns of York and Dublin, all locations where there were no local sources of steatite. Together with the evidence from Norway and Shetland, a picture of the use of steatite in the North Atlantic has emerged. As Forster suggests, the Landnám phase is characterized by the transport of original imports from Norway with the migrants when they moved to the North Atlantic islands. The origins of the steatite artifacts are also almost exclusively Norwegian. The only exceptions are the baking plates recovered from Old Scatness, which do not appear in Norway until the 12th century, together with the less wellmanufactured vessels from Shetland. A typical asJournal of the North Atlantic 9 2016 No. 29 J. Marttila is possibly the result of sporadic exchange or contact between the elites of the island groups (Forster 2004:360–363). The control of quarries in also evident in Norway where areas such as Hyllestad may have been under the direct control of local magnates during the Viking period and by the 12th century were controlled by non-local political and economic elites. Tenant farmers appear not to have been able to use quarries for their own benefit. Eventually the lands were transferred to ecclesiastical institutions (Baug 2015:141). Fishing The artifactual, ecofactual, and, to some extent, historical evidence indicates that fishing was an important part of the economy in Shetland during the period of Scandinavian settlement. The evidence indicates that both line and net fishing were practiced. However, there is a considerable variation of the quality of the assemblages from different sites, depending on the excavation methods and the conditions for preservation (Graham-Campbell and Batey 1998:214). Evidence from Shetland Every Norse excavation in Shetland has provided artifactual evidence of fishing. For example, the excavations at Underhoull have recovered artifacts inclusive of small line sinkers and a sandstone boulder probably used for sharpening fishing hooks. Fish were probably also used for oil production at the site, as 2 artifacts, unique to Underhoull, appear to have been used for rendering down fish livers (Small 1966:241–242). Artifactual evidence of fishing was recovered from the excavations at Sandwick South and Jarlshof as well. Line-sinkers were recovered from both of the sites, and Jarlshof even produced evidence of fishhooks. Sandwick South also provided evidence of linesinker production, as sinkers in various stages of manufacture were recovered from the living quarters of the dwelling (Bigelow 1985:119, Hamilton 1956:153). The ecofactual assemblage, however, provides the most extensive evidence of the Norse fish exploitation and its effect on the economy during the Viking and late Norse periods (Graham-Campbell and Batey 1998:214). Unfortunately, sites such as Underhoull and Jarlshof were excavated prior to the introduction of modern methodologies, and thus, the recovered assemblages from those sites are limited. The excavations at Old Scatness provided extensive evidence of fishing, although a relatively limited range of species were present. Gadids, in particular remain extremely few and are exclusively from Norway. The low numbers of Norwegian goods throughout the North Atlantic suggests that steatite could have only been a secondary trade good and therefore linked with the growing stockfish trade and the movement of staple goods in the North Atlantic (ibid.:327–330, 359; see above). The analysis of the origins of steatite is at the moment largely based on typology. However, emerging techniques such as rare-earth element analysis by ICP-MS (see Jones et al. 2007) can be used in the future to determine the origins of the artifacts and confirm/refute the typological theories. The control of the steatite resources in Shetland As mentioned above, the exported vessels from Shetland appear to have originated from only Catpund, Cunningsburgh, and the vessels produced were of a standard form. Therefore, it is highly likely that the production and distribution of the quarry products were controlled at least by the end of the 10th century (Forster 2004:330, 362). However, this is not necessarily the case for the other, smaller-scale quarries in Shetland, where the production could have been for local consumption and for domestic use. It is possible, however, that some farms specialized in steatite vessel production and produced a surplus of vessels which they could have traded locally. Evidence from Belmont, Unst, possibly suggests the theory, as the final stages of vessel production took place at the settlement (Larsen 2013:181). Evidence from Catpund, Cunningsburgh (Forster 2004), suggests that individuals could have been allocated areas within the quarry that they could utilize. This, again, is a strong indication of what is termed “resource control”. Further indication of the possible increase of control is suggested by evidence from sites such as Sandwick South (Bigelow 1985), which are located close to steatite outcrops but have not produced any evidence of manufacturing. It is possible that these sites were not given permission to participate in steatite manufacturing and had to rely on other resources. Despite the extensive exports to Orkney, the spread of Shetlandic steatite products appears to have been limited within the political sphere of the Earldom of Orkney. It is possible that the resource control in Shetland is related to the development of the Earldom, and the distribution of the commodity therefore reflects its territories. The continuation of the trade beyond the separation of the Earldom in 1195 suggests that links between the archipelagos did not cease to exist. The recovery of the Shetland type of vessels from the Western Isles and the Faroes 2016 Journal of the North Atlantic No. 29 J. Marttila 10 assemblage, which possibly indicates the production of cured fish for exchange (Barrett 1997:618; see also Perdikaris 1996, 1998). The ecofactual assemblage from Jarlshof covered both the Viking Age and the Late Norse period. Bones of cod, ling, and saithe were recovered (Hamilton 1956:214). However, if modern excavation techniques were used, the amount of recovered fish bones would have most likely been significantly greater. According to Bigelow (1985:121), the contrast in fish-bone levels at Jarlshof and Sandwick fits the classic pattern of under-representation of small bones. In any case, the (seemingly partial) evidence from Jarlshof suggests that fishing became more important in the Late Norse period as, according to Hamilton (1956:157), steatite line sinkers appeared in “ever-increasing numbers”. Moreover, all the fishhooks were recovered from the later contexts as well (ibid.:153). The dominant species during all periods at Old Scatness was saithe, which is a reflection of the abundance of the species in the waters around Sumburgh Head. Small saithe could have easily been obtained by fishing from the shore. The Viking contexts from Structure 6 were dominated by larger cod and saithe. This structure may have functioned as a store/ smoke house (Dockrill et al 2010:165–177). The late Norse evidence from Old Scatness was recovered from a midden situated northeast of the Iron Age village. The midden provided an extensive assemblage of fish remains. Compared to the previous periods, the proportion of cod to other fish increased and the fish became more substantial in size (Turner et al. 2005:247). These larger cod, ling, and saithe must represent boat-based fishing, and some may have been processed as stockfish (Dockrill 2010:167). Trade in fish products Historical sources describe the emerging trade in cured fish in northern Europe, albeit only to a limited extent. The first possible, although somewhat unreliable, indication is an anecdote in the Egil’s Saga (written in the first half of the 13th century) that mentions fish exportation from Norway to England in the 9th century (Pálsson and Edwards 1978:49). By the 11th century, the evidence becomes more reliable, and by the early 12th century the evidence is strong. For example, the records from this century indicate that a person fishing in Vågan was required to pay 5 fish as a tax by a royal decree during the combined reign of the Kings Eysteinn, Sigurðr, and Óláfr Magnússon. Slightly after this time, houses and a church were ordered to be built for the fishermen in the area (Urbanczyk 1992:132–133). saithe (Pollachius virens), was the most dominate species. The nature of the contexts from which the fish-bone assemblages were recovered varied greatly though. The Pictish and Viking evidence derived primarily from domestic structures, and the late Norse evidence was excavated mainly from large coastal middens (Dockrill et al. 2010:165–167). The Viking evidence consists mostly of young saithe that were most likely caught from shore using nets or rods and line. Mature saithe, cod, and ling were abundant in the Norse middens, being indicative of boat-based fishing. The evidence suggests that some of the large fish were also processed as stockfish. It remains unclear, though, whether the stockfish was consumed locally or exported elsewhere (ibid.:165–167). The fishbone assemblage from Sandwick South was dominated by young saithe of lengths under 50 cm, suggesting that they were most likely caught from the shore. However, large fish from the cod family were also recovered, which must represent linefishing from boats. The bones were deposited outside in the yard, often in substantial concentrations. These fish were crucial for the daily subsistence in Shetland, although oil could have also been produced for trade from them (Bigelow 1985:191–193). Shellfish (both limpets and whelks) were also recovered in considerable quantities from Sandwick South (ibid.:121). A substantial late Norse shell deposit was also recovered from Old Scatness. However, it remains uncertain whether at Old Scatness shellfish were used as bait for fishing or represent culinary refuse (or a mixture of both) (Dockrill et al. 2010:176–177). Limpets were often used as fishbait in later times, and historical records indicate that both limpets and whelks were eaten during periods of famine (Fenton 1978:542). Even though they are poor as a long-term food resource due to their low nutritional value, they were crucial in northern maritime environments as a short-term source of calories and assured the survival of a population, as they were available in abundance throughout the year (Bigelow 1985:121). The substantial amount of fishbones and shellfish indicate the importance of the marine environment as a source of food at Sandwick South during the Late Norse period. Bigelow (1985:123) interpreted the small saithe as representing subsistence-based activities. The larger specimens would have been essential to a commercial-orientated fishery, but “the faunal collections from an archaeological site … cannot supply evidence for that sort of specialization” (ibid.:123). However, an imbalance in the ratio of cranial and tail elements was identified in the Journal of the North Atlantic 11 2016 No. 29 J. Marttila constituted 97.9 percent of the total dry bone weight. Cod, ling and saithe were the most common species (Barrett 1992:10, 23). Barrett (1997:621) suggested a model for cured fish production, according to which the exported gadid fish probably contained the bones of the appendicular skeleton and some or all of the vertebral column. The cranial bones and the anterior portion of the vertebral column possibly remained at the processing sites. He also identified possible complication in the pattern, as for example fish heads may have been discarded at sea, used as animal fodder, or spread as manure. Also the fish that spoiled during the processing were probably discarded to the middens, as well as the bones of locally consumed fish. Furthermore, the variation in preservation has an effect on the recovered assemblages (ibid.:622). Despite the potential for uncertainty, the overall evidence from Robert’s Haven suggests that cod, saithe, and ling were processed according to the proposed cured-fish model: cleithra and other appendicular bones were under-represented, transverse cut marks indicate that the gadid were often decapitated anterior to the cleithrum, the posterior caudal vertebrae were under-represented, and butchery marks indicate that only the anterior portion of the spine was cut off during processing (ibid.:628). These characteristics are consistent with zooarchaeological evidence recovered from other Norse coastal sites in Iceland and the Faroes (see Perikaris et al. 2007) Earl’s Bu, Orkney. Earls Bu is located on the south of Mainland Orkney, approximately 200 m inland from the shore of Scapa Flow. The site consists of a Viking Age horizontal mill which was superimposed by middens associated with the Medieval elite settlement located in the vicinity (Batey 1993, Batey and Morris 1992). Parts of the midden, which dated from the 11th and 12th centuries, were studied by Barrett (1997:628–629) for comparison with the Robert’s Haven assemblage. The midden at Earl’s Bu had a domestic character and is thought to have originated from the occupation of the nearby Bu (an elite settlement). Its bone preservation was poorer than the middens at Robert’s Haven. Despite this, 13 different species of mammals were identified from the bone assemblage. Fish bones were also abundant and were dominated by the gadids (ibid.:629, Batey and Morris 1992:34– 36). Cod (39.1 percent) and haddock (22.4 percent) were dominant in the assemblage, but saithe (3.3 percent) and ling (2.1 percent) were almost absent (Barrett 1995:262). The genuine abundance of haddock was clearly identifiable, even though the Bergen, strategically situated between the fishing districts of the north and the Continental European markets, gained a dominant position in the export of stockfish, fish oil, and other fish products to areas around the North Sea and the Baltic, especially during the Hansa era. Zooarchaeological evidence from Lofoten, northern Norway, indicates that around the 13th century there was a significant increase in stockfish production. Perdikaris (1999) suggests that this was a result of the economy changing from the local chieftains trading for prestage to the international market economy under ecclesiastical and royal control. Iceland appears to have become involved during the late 13th century, when the fishing became more specialized, as was the case particularly in the islands of Vestmannaeyjar in the south and the peninsula of Snæfellsnes in the west. The chieftains and ecclesiastical institutions soon gained control of the best fishing farms and settlements and as a result controlled the export of fish products, initially via Bergen (Helle 2003:284). During the second half of the 13th century, the German Hanseatic merchants gained control of the Norwegian markets and expanded it to include (for example) the important trade between Bergen and England. By the 14th century, they also controlled northwestern continental Europe, the Rhine regions, and the Baltic (ibid.:284). The earliest evidence of Hanseatic merchants visiting Shetland is from ca. 1415 (Friedland 1983:88). However, the references to Scandinavian Scotland during the earlier centuries are sparse and therefore do not indicate from what period onwards Shetland was incorporated into this European trade network (Barrett 1997:616). As the historical evidence for the fish trade in the Earldom of Orkney is sparse, the archaeological evidence can provide significant additional information of the fish trade in the Northern Isles. However, as no substantial fish-rich middens have been investigated with modern methodologies in Shetland, evidence from Orkney and Caithness will be used in the discussion instead. Barrett (1997) assessed the zooarchaeological evidence from Robert’s Haven in Caithness and Earl’s Bu in Orkney in order to determine whether the production of cured fish was identifiable in the archaeological records as a bulk/ staple item. Robert’s Haven, Caithness. A coastal, fish-rich midden was analyzed at Robert’s Haven that measured approximately 25 m in length and 1.2 m in depth and was dated to the 13th/14th centuries. Fish bones dominated the recovered bone assemblage and 2016 Journal of the North Atlantic No. 29 J. Marttila 12 a product of seasonal inshore and offshore fishing, which probably also generated surplus. The surplus could have been traded locally in Caithness or in Orkney in return for agricultural products, or used to pay rent or taxes (Morris et al. 1995:269). Barrett (1997:634) identified 7 factors that suggest stockfish exports from the Earldom of Orkney: • There is evidence of long-distance trade from the earldoms in the form of imported products and (limited) historical evidence. • The fish-processing evidence from Robert’s Haven is consistent with the production of cured fish products, which were traded in northwestern Europe between the 14th and 16th centuries. • The intensity of fishing in Robert’s Haven is also consistent with Post-medieval commercial fishing stations in Shetland. • There are also other middens that possibly indicate the same intensity as Robert’s Haven (e.g., Quoygrew, St. Boniface, and Sandwick South). • Circumstantial historical evidence from the 12th century possibly indicates that the earldom participated in stockfish trade. • A record survives which mentions that the Royal Exchequer of Scotland bought 15,000 dried fish from Caithness in 1329. • The pattern is consistent with 15th-century and later records concerning Orkney, Shetland, and Caithness that specifically discuss the fish trade. An intensification of fishing has also been identified from some sites during the Late Norse period. For example, more fish were recovered from deposits in Buckquoy dating to Late Norse phases IV and V than from all the preceding phases together (Graham-Campbell and Batey 1998:215). The artifactual assemblage from Shetland might suggest intensification as well, as there is an increase of imported goods from the 12th century onwards and, therefore, exportation probably increased as well, most likely of dried/stock fish (Bigelow 1989:190). This increased need for dried fish has been explained by factors such as Lenten fare or military rations (Barrett 1997:616, Graham-Campbell and Batey 1998:218). It is likely that the Shetlandic situation was similar to the one in Orkney and Caithness, and similar changes in fishing practices probably occurred during the Late Norse period. It is, therefore, possible that similar archaeological deposits to the ones identified at Robert’s Haven, Freswick Links, and Earl’s Bu could be discovered in Shetland when more coastal sites are excavated. species is often exaggerated on sites with poor bone preservation. As haddock is a deep-water taxon that is not currently present in the vicinity of Scapa Flow, it is possible that they were caught to the west of Orkney or in the Pentland Firth. Also of interest is that saithe, which is typically the second-most abundant taxon in the Viking Age and Medieval assemblages from the Earldom, was practically absent at Earl’s Bu. Due to the settlement’s high status, it is possible that the inhabitants controlled resources from throughout the earldom and therefore were in a position to exclude the fish from their diet (Barrett 1997:629–631). The zooarchaeological assemblage suggested that decapitated fish had been transported to the site. In addition, cut marks identified from fish bones, although few in number, are consistent with the consumption of cured fish at the site (ibid.:632). Freswick Links, Caithness. An excavation was carried out at Freswick Links between 1980 and 1982 that concentrated on midden deposits eroding at the coast. During the excavation, 40 tons of archaeological deposits were wet-sieved with 1-mm mesh. The fish bones were generally well preserved (Morris et al. 1995:154–155). The assemblage was dominated by large gadid fish, but concentrations of small saithe were also identified. However, the fish bones were not evenly distributed within the excavated area. The eroding middens at the cliff edge were dominated by fish bones and shells, but mammal bones were more common than fish bones in the middens further inland (Graham-Campbell and Batey 1998:216, Morris et al. 1995:154–155). The evidence from the late Norse deposits suggested activities such as cooking, fish drying and smoking, and bait preparation (Morris et al. 1995:268–269). Thousands of fish were represented in the Freswick links middens, and in some areas of the middens over 100 fish were processed at the same time. Even then, only 1 area produced an assemblage substantial enough to be interpreted as waste products of commercial-scale fishing. According to Morris et al. (1995), if the stockfish trade had been an important part of the community’s economy (for example, similar in scale to the Norwegian stockfish trade), there should have been more of the large assemblages. The fact that imported products cannot be easily detected also suggests that large-scale export of stockfish might not have occurred at Freswick (Graham-Campbell and Batey 1998:218). However, even if fishing was not commercial in scale, the assemblages indicate the significance of it to the local community. The middens could be interpreted as Journal of the North Atlantic 13 2016 No. 29 J. Marttila Whales, Seals, and Birds Whales Whales were hunted by the Norse for their meat, oil, blubber, and bones. The whales were caught by the Norse with several methods, of which the 2 most common ones appear to have been the catching of large whales by hand-harpooning from boats and the driving and beaching of long-finned pilot whales. The former was still practiced in Norway in the Late Medieval period and continued sporadically in Iceland into the 18th century (Lidquist 1997:7–8). The latter is still regularly practiced in the Faroes, where the catches have been recorded since 1584 and still constitute 30 percent of the archipelago’s meat production (Department of Foreign Affairs, Faroe Islands 2007). The method was also practiced in Shetland and Iceland and to a lesser extent in Orkney, the Western Isles, and Ireland (Lidquist 1997:29). Drive-beaching hunts occurred when a pod of whales was discovered in close proximity to land and when the weather conditions were suitable. The Faroese customs suggest that a number of boats gathered in a semicircle behind the pod and drove the whales towards a chosen firth. When the whales were close to the beach, a hook with a rope attached to it was driven into the backs of them, and they were pulled to the shore and killed by cutting the spinal cord. The catch was divided and the local occupants received an equal share of the meat and blubber (Department of Foreign Affairs, Faroe Islands 2007). The whole community today is involved in the event. In Shetland, the place-names—such as Whale Firth in Yell, and Whalegarth and Whalayre in Unst—are a clear indictor of whale hunting. Whale bones have also been identified from Norse contexts from (for example) Jarlshof and Old Scatness (Dockrill et al. 2010:136–137, 156; Hamilton 1956:214). The practice appears to have continued well into the Post-medieval period, as an account from the early 1800s narrates that a pod of ~200 whales had entered the Uyea bay and the locals prepared to catch them, but had to wait due to the Sabbath and, as a result, the whales escaped (Atkinson 1832 cited in Hunter 1994:262). Seals Seals were hunted by the Norse as well for meat, blubber, and bone. They were usually taken in groups by clubbing or netting at their habitual gathering places (Fenton 1978:26). Evidence from Jarlshof indicates that seals (especially grey seals) were commonly hunted, particularly during the 10th and 11th centuries, when they were utilized as much as domestic animals (Platt 1956:215). They have been identified from both the Viking and Late Norse contexts from Old Scatness (Dockrill et al. 2010:136–137, 156) and the Sandwick South assemblage as well (Bigelow 1985), but only in traces. The lack of utilization at Sandwick South is possibly due to the settlement not having access to the required resources if the seal culls were strictly controlled by the landowners during the Viking and Late Norse periods as has been documented for the Post-medieval period. Interestingly, by 1360, the 2 important sealkilling locations in southern Unst, Uyea and Haaf Grunay, were controlled by the powerful landlords of Papa Stour. It is possibly a coincidence, but could also be an indication of the control of the resource already at the time (Ibid.:120). Birds Both Jarlshof and Sandwick South also provided evidence of the utilization of birds. Bird bones were identified in the faunal collections at Sandwick, but not in abundance and in far lesser quantity than the remains of fish (ibid.:121). There is an abundance of seabirds around Jarlshof in the present day, and both domestic and wild bird bones were identified in the bone assemblage (Platt 1956:214). The bird remains recovered from the Viking and Late Norse contexts from Old Scatness were relatively sparse as well. This can be seen as an indication that birds constituted a fairly minor food resource. The species eaten included gulls, shag, cormorant, and gannet (Dockrill et al. 2010:171). Animal Husbandry and Cultivation The bone assemblages and the presence of byres indicate that cattle had a significant role in the Norse economy of Shetland. Sheep also appear to have been important for both meat and wool production. For example, Sandwick provided evidence of butchered sheep bones, and loomweights and spindle whorls have been recovered from every Norse excavation within the research area. Both botanical and artifactual evidence of the use of grain were also recovered from Sandwick, and several rotary querns and carbonized seeds of hulled barley and oats were recovered in the excavation. The recovery of charred seeds from several excavated areas suggests local cultivation rather than importation (Bigelow 1985:118–189). Hamar and Belmont also provided evidence of cultivation, as both hulled six-row barley and oats were recovered during the excavations. Interestingly, some of the barley grains appeared to have been 2016 Journal of the North Atlantic No. 29 J. Marttila 14 harvested before they were fully ripened. The probable explanation for this is the challenging environment and short growing season in Shetland (Turner et al. 2013:146, 209). Flax/linseed was also recorded in several samples from Belmont, and it appears that it was processed at the site (Ibid.:209). The excavation at Upper House, Underhoull, also provided evidence of cultivation. Barley, oat, flax seeds, and a single grain of what is likely bread wheat were recovered. It is likely that the recovery of barley, oat, and flax represent cultivation on the site. It is unlikely, however, that wheat was cultivated due to the difficult environment. It could have been present as a weed or imported through trade networks. Overall, the density of cereal from the site was low, which in itself might be an indication that cereals were not used with great intensity at Upper House, or, alternatively, that the cereals were imported from elsewhere (ibid.:172). Artifactual evidence from Underhoull is also suggestive of cultivation, as 2 rotary querns and broken points of ploughshares were recovered during the excavation (Small 1966:239–240). At Old Scatness during both the Viking and Late Norse periods, by far the most economically important species were cattle, sheep, and pig. Cattle were primarily used for dairy production, although there is evidence of beef consumption as well. Sheep were utilized for wool, meat, and possibly skin. Pigs were used almost exclusively for their food production. Horse, dog, and cat bones were also present suggesting that they would have been utilized for meat or skins, although their main purpose would likely have been as live work animals (Dockrill et al. 2010:136–137, 156). The main cereal during the Viking and Late Norse periods at Old Scatness was six-row hulled barley. Oat, however, appeared to increase in importance during the Norse settlement and is present in almost 80% of samples from the Viking period and in all Late Norse samples. It appears that the Norse also introduced flax to Old Scatness, which would have provided linseed oil and flax fiber (ibid.:193–195). Jarlshof provided evidence of animal husbandry as well, although as it was excavated over half a century ago, the techniques of recovering organic materials and environmental data were poor. As a result, the report on organic remains consists of just one page for the whole site and the report of animal bones from the Norse deposits consists of only 2 pages. The report indicates that sheep, ox, pig, horse, seal, red deer, dog, whale, fish, and bird bones were encountered during the excavation. The sheep bones were common during the whole settlement period, but mainly consisted of older animals, which would suggest that the secondary products were more valued than the meat (Hamilton 1956:214). This finding might partly be due to the taxation in Shetland, as the principal tax commodity during the medieval period was butter. The intensification of dairying is also suggested by the increase of the early mortality of calves during the Late Norse period at Sandwick South (Bigelow 1989:188–189). According to Hamilton (1956:214), oxen were common at Jarlshof, and pig remains consisted of both young and old animals. Pigs were expensive to raise in the northern environments and could be very destructive to cultivated areas and increase the risk of erosion. It is also possible that smaller settlements such as Sandwick South could not afford them (Bigelow 1985:120). Imported Commodities Every excavated settlement in Shetland has provided evidence of imported commodities. Consequently, it is obvious that the archipelago had connections to other lands in Scandinavian and even further afield. Shetland offered a staging post on several seaways: to Iceland and beyond; to the Western Isles and Ireland; to the Fair Isle, Orkney, and the British east coast; and, of course, to Scandinavia (Morrison 1973:384). The imported materials indicate that settlements in Shetland did not exist in isolation but were part of a lar ge economic system. Sandwick South provided evidence of imported pottery which included English Scarborough and Grimston pottery and North German proto-stoneware from phases 2–3 (AD 1200–1350). It is likely that all the bone combs recovered from the site were imported as well, as no manufacturing waste was identified and very similar combs have been recovered in Norway. In addition, an hourglass-shaped lamp was discovered that has close parallels from Norwegian urban sites from the 13th and 14th centuries (Bigelow 1985:108, 121–124). A similar lamp was discovered from the Biggings, Papa Stour, as well (Crawford and Ballin Smith 1999:139). It is possible that the economy at Sandwick and elsewhere in Shetland underwent significant changes at some point and the major marine resources were used for exportation as well as consumption (Bigelow 1985:108, 121–124). The artifactual evidence from Jarlshof also clearly indicates the site’s far-reaching connections. The evidence from combs show clear links to Scandinavia. During the Late Norse period at least, combs were produced in the Scandinavian manufacturing centers and exported widely around the Atlantic. The combs recovered from Jarlshof follow the stylistic Journal of the North Atlantic 15 2016 No. 29 J. Marttila developments in Scandinavia (albeit with a possible delay). For example, the comb found from a Phase II midden, probably of mid-9th-century style, has close parallels from the Oseberg burial (Hamilton 1956:130). The double-sided “fish-tail” composite comb recovered from a Phase VII context has close parallels from Norway, Denmark, and Sweden (Clarke and Heald 2002:83). Further links can be demonstrated with metalwork. For example, a trefoil mount that Hamilton (1956:150) interpreted as a brooch found from a Phase III (10th-century) context has close parallels from Iceland (2), North Yorkshire, and Lincolnshire. The mounts are of insular style and were most likely manufactured by the same master somewhere in the British Isles (Paterson 1997:649–656). A significant amount of other artifacts recovered from Jarlshof have their origins elsewhere as well. Some of the steatite vessels might have originated from Norway, for example (see above). The Biggings provided evidence of extensive importation of different materials, which is an indication of the great wealth of the settlement. The importation is of a scale rare for rural settlements of the Early and High Middle Ages in Scotland. The quantity and variety of imported goods increases from the 11th and 12th centuries until the beginning of the 17th century. The most dominant group of imports is the ceramics. The imported wares include, for example, unglazed Paffrath ware, proto- and near-stonewares from Germany, Medieval English wares, and North German and South Scandinavian redwares (Crawford and Ballin Smith 1999:99, 127). The significant amount of wood recovered from the Biggings is a clear indication of trade as well. The imported wood was mainly used in construction of the buildings from the 11th to 15th centuries, but also for the manufacturing of small artifacts, such as spoons. Historical evidence—along with the size of the wood used, and the quantity of Scots pine and oak timber—also suggests that the wood mainly originated from southern or eastern Norway. It seems that bark and branches were removed from the timbers prior to the importation to the site. The evidence from the wooden floor indicated that the timbers had been radially split and prepared prior to the importation as well. Some of the smaller items were probably also imported, as their manufacturing required significant amount of skill, which the people living in treeless environment probably did not have. Other imported wood includes beech and ash probably from Scandinavia, cork oak from the western Mediterranean, and balsa wood. The occurrence of balsa wood in a context dating to 13th/14th centuries remains unexplained as it grows in Caribbean and tropical America. The total lack of marineboring molluscs infiltrating the timber suggests that only a trivial amount of the wood recovered is driftwood (ibid.:79, 98–99, 127). Imported items were also recovered from Old Scatness, the bulk of which were steatite vessels and artifacts (Dockrill et al. 2010:266–288). Pottery was also imported which included possibly Late Roman grey-ware sherds and a Middle Saxon import from the east coast of England (ibid.:227–228). Other imports include several glass beads (ibid.:347–348). Furthermore, evidence of extensive metalworking at Old Scatness suggests a sophisticated metal trade, including tin or tin bronze from southwest Britain as, due to the nature of the metalworking, it cannot have been sustained by an economy based on scrap metal (ibid.:344–345). Specialization of the Settlements during the Late Norse Period It appears that the settlements began to specialize their production during the Late Norse period. As seen above, the evidence from steatite production and fishing both suggest it to some extent. The excavation at Belmont provided evidence of steatite production (Larsen 2013:181), which was not identified at Sandwick, for example (Bigelow 1985). The specialization is also suggested by the distribution of the Late Norse settlements. As the location of the settlements expanded into more peripheral areas without access to the sea, it would not have been feasible for them to take part in fishing as an everyday activity. For example, Snabrough is located so far inland that it would have taken too long to get to the shore and back every day (Fig. 7). Economic specialization is also suggested by the place-name evidence. For example Niddister, Northmavine (first mentioned in 1490), from where there is no easy access to the sea, derives from Old Norse naut, n., cattle and setr, a farm (Stewart 1987:229, 234). This nomenclature strongly suggests that the settlement specialized in cattle husbandry at the time of its coining. Another example is Smirgarth, Unst. The settlement’s name derives from the Old Norse Smjör-gerði, from smjör, n. butter gerði, a fenced piece of land (common in farm-names) (ibid.:91, 94). It is highly likely that the site focused on butter production at the time of its coining. Other sites whose place-names suggest specialization within the research area include Bighton, Unst (from bygg, n. a barley), Hestaness, Fetlar (from hestr, a stallion or horse), Collaster, Unst and Colvister, Yell (from kálfr, m. a calf), Bixsetter, Yell (from bygg, n. 2016 Journal of the North Atlantic No. 29 J. Marttila 16 Figure 7. Norse Settlements in Northern Shetland. Journal of the North Atlantic 17 2016 No. 29 J. Marttila barley), Buster, Yell (from bú, n. cattle), Russetter, Fetlar (from hross, n. a horse or mare), Sodasness, Cuppister, Yell (from sauðr, m. a sheep), Swinaness, Unst (from svín, a pig), Buness, Unst (from bú, n. cattle), Priesthoulland, Northmavine (from prestr, m. a fish), Sütwell, Northmavine (from sauðr, m. a sheep), Bool, Unst (from bú, n. cattle), Sweinsgarth, Yell (from svín, a pig), Clipprigarth, Unst (from klippari, m. a sheep shearer), Grisgarth, Unst (from gríss, m. a hog, young pig), Colvadale, Unst (from from kálfr, m. a calf), Cubil, Fetlar (from kýr, f. a cow), and Ockran and Ekren in Northmavine (from akr, m. or ekra, f. a cornfield) (ibid.:36, 73, 93, 96, 98, 140, 142, 212, 228, 230, 231, 272). As these references to specific elements of farming are so common, they can be seen as an indication of a large-scale specialization in the community. It is also worth noting the substantial variety of different products indicated by the place-names. They suggest that some farms focused on animal husbandry and others on cereal production. Farms with an easy access to the sea continued with fishing as their main economic activity. This specialization was probably at least partly dictated by the pressure on the land and from the resulting habitation of inland locations which were not preferable for conducting fishing as a primary production activity. In these circumstances, the resources available for the settlements began to vary. Some farms would have had a more substantial space available for them suitable to carry out various forms of animal husbandry or cultivation. Depending on the location, the farms would have chosen the type of animal or crop suited to their particular environment. As noted, the place-name evidence suggests that at least one farm even may have specialized in butter production. The evidence also indicates that at least some farms engaged primarily in the cultivation of barley and corn. Also as a result of specialization, the settlements probably would have created a surplus of their particular products that would have been sold or traded in return for other commodities. For example, the coastal farms could have traded fish, together with fishing by-products that could have been used as fertilizer for infields, to the inland settlements. Control of the resources could have influenced settlement specialization as well. As mentioned above, it is possible that seal culls were strictly controlled by the time of the Viking and Late Norse periods and that some elite settlements, such as the Biggins, could have had exclusive access to important seal-killing locations. It is possible that the control of steatite outcrops and production influenced specialization. Evidence from Catpund suggests that areas of the quarry were utilized by different individuals, which could be seen as indicative of control. Also, the uniformity of steatite vessels exported from Catpund to Orkney and, to a lesser extent, the Western Isles and the Faroes suggests control of the resource (Forster 2005:69). Furthermore, Belmont has produced evidence of the latter stages of vessel manufacturing even though there are no outcrops in its vicinity. These findings strongly suggest that steatite was a highly controlled resource and only farms/individuals with permission could utilize it. It is therefore likely that these farms/ individuals produced a surplus that was traded for other commodities. The uniformity of the steatite vessels also suggests commoditization of the object being traded. Uniform objects would have increased the buyers and sellers understanding of the objects’ value, utility, and other characteristics. This combination of evidence indicates that the economy of late Norse Shetland appears to have been a mix of subsistence and trade. Further evidence of specialization could be gathered by excavations of the farms and their associated middens together with mapping and comparing their infield and the surrounding field boundaries. Summary Even though early historical sources indicating Shetland’s connection to the wider Viking world are limited, archaeological evidence clearly indicates that Shetland was integrated into the North Atlantic network, which consisted of both rural settlements as well as centralized places for exchange on the numerous islands, archipelagos, and lands. All excavations have provided evidence of acquisition of imported artifacts and, in some cases, also of bulk goods (see the Biggins above). Shetland provided a staging post on several seaways in the North Atlantic: to Iceland and beyond; to the Western Isles and Ireland; to Orkney and the east coast of Britain; and to Scandinavia proper. Archaeological evidence also clearly indicates that the inhabitants of the archipelago used a variety of local resources for both subsistence and for market production. Due to Shetland’s circumstances, fishing played a crucial part in the economy of the islanders, but the sea also produced other food sources such as seals, whales, and seabirds, all of which were utilized by the islanders. The land available for the inhabitants also was used for a variety of resources. The evidence indicates that cultivation played an important part together with animal husbandry. Steatite production was also important at least for some of the settlements. 2016 Journal of the North Atlantic No. 29 J. Marttila 18 Evidence from steatite quarries and settlements suggest that the production at the larger quarries, such as Catpund, was controlled to some extent. It is also possible that other resources, such as seal colonies, were being controlled as early as during the Viking and Late Norse periods by the larger landowners. Archaeological and place-name evidence suggests that at some point during the Norse settlement of Shetland the settlements began to specialize. It appears that some focused on fishing, others on steatite production, and others on different types of animal husbandry and cultivation in addition to their subsistence production. As a result of this specialization, the farms would have begun to produce surplus of their particular product and traded this for other products locally or, in some cases, possibly even outside of Shetland. The economy in Shetland therefore began to develop from a purely subsistence- based economy towards a market economy during the early second millennium. Acknowledgments This research was carried out for my Ph.D. thesis at University of Glasgow. I would like to thank my Ph.D. supervisor, Dr. Colleen Batey, for her assistance and guidance. I would also like to thank Dr. Julie Bond and Steve Dockrill for their advice and allowing me to participate in, and access the information from, several excavations in Shetland; Val Turner and Carol Christiansen for providing me with information from the databases of the Shetland Amenity Trust and sharing their knowledge of the area; Beverley Ballin Smith for providing me information of the Norwick excavations; Brian Smith for providing me information of the historical records of Shetland and the Shetland Archives; Ian Tait for allowing me to access the collections of the Shetland Museum; Zoe Outram for providing me with reports from the excavations and surveys in Shetland; Dr. Terence Christian, University of Glasgow, for his assistance with the illustrations; Simun Arge for sharing his expertise of the Norse in the Faroes; Anne-Christine Larsen for her information of the Belmont excavation; Dr. Tom Horne for his assistance; and the Osk, Huttusen Säätiö, for funding the research. Literature Cited Arge, S.V. 2000. Vikings in the Faeroe Islands. Pp. 154–163, In W.W. Fitzhugh and E.I. Ward (Eds.). 2000. Vikings: The North Atlantic Saga. Smithsonian Institution Press, Washington, DC, USA. Atkinson, G.C. 1832. An Excursion to the Shetland Isles. Manuscript. [WHERE AVAILABLE FROM?]. Barrett, J.H. 1992. Palaeoeconomic Investigations at Robert’s Haven, Caithness, 1992. A preliminary report. University of Glasgow, Department of Archaeology, Glasgow, UK. Barrett, J.H. 1995. “Few know an earl in fishing-clothes”: Fish middens and the economy of the Viking Age and Late Norse earldoms of Orkney and Caithness, Northern Scotland. Ph.D. Thesis. University of Glasgow, Glasgow, UK. Barrett, J.H. 1997. Fish trade in Norse Orkney and Caithness: A zooarchaeological approach. Antiquity 71(273):616–638. Barrett, J.H., R. Beukens, I. Simpson, P. Ashmore, S. Poaps and J. Huntley. 2000. What was the Viking Age and when did it happen? A view from Orkney. Norwegian Archaeological Review 33(1):1–39. Batey, C.E. 1993. A Norse horizontal mill in Orkney. Review of Scottish Culture 8:20–28. Batey, C.E., and C.D. Morris. 1992. Earl’s Bu, Orphir, Orkney: Excavations of a Norse horizontal mill. Pp. 33–42, In C.D. Morris and D.J. Rackham (Eds.). 1992. Norse and Later Settlement and Subsistence in the North Atlantic. University of Glasgow, Glasgow, UK. Baug, I. 2015. Quarrying in Western Norway. An Archaeological Study of Production and Distribution in the Viking Period and Middle Ages. Archaeopress Publishing Ltd, Oxford, UK. Bigelow, G.F. 1985. Sandwick, Unst, and Late Norse Shetland economy. Pp. 95–127, In B. Ballin Smith. (Ed.) Shetland Archaeology. The Shetland Times, Lerwick, Shetland. Bigelow, G.F. 1989. Life in medieval Shetland: An archaeological perspective. Hikuin 15:183–92. Bigelow, G.F. 1992. Issues and prospects in Shetland Norse archaeology. Pp. 9–32, In C.D. Morris and D.J. Rackham. (Eds.). Norse and Later Settlement and Subsistence in the North Atlantic. University of Glasgow, Glasgow, UK. Bray, I., A. Forster, and S. Clelland. 2009. Steatite and Shetland: A geological introduction and gazetteer of sites. Pp. 4–17, In A. Forster and V.E. Turner (Eds.) 2009. Kleber: Shetland’s Oldest Industry. Shetland Soapstone since Prehistory. Shetland Amenity Trust, Lerwick, Shetland. Buckland, P.C., and J. Sadler. 1990. Ballast and building stones: A discussion. Pp. 114–125, In D. Parsons (Ed.) 1990. Stone: Quarrying and Building in England AD 43–1525. Phillimore, London, UK. Buttler, S. 1983. Cross Geos, Clibberwick (Unst): Discovery and excavation in Scotland 1983:21–22. Buttler, S. 1989. Steatite in Norse Shetland. Hikuin 15:193–206. Clarke, D., and A. Heald. 2002. Beyond typology: Combs, economics, symbolism, and regional identity in Late Norse Scotland. Norwegian Archaeological Review 35(2):81–93. Crawford, B.E., and B. Ballin Smith. 1999. The Biggings, Papa Stour, Shetland. The History and Excavation of a Royal Norwegian Farm. Society of Antiquaries of Scotland and Det Norske Videnskaps-Akademi, Edinburgh, UK. Department of Foreign Affairs. Faroe Islands. 2007. Whales and whaling in the Faroes. Available online at http://www.whaling.fo/thepilot.htm. Accessed 27 November 2007. Journal of the North Atlantic 19 2016 No. 29 J. Marttila Dockrill, S.J., J.M. Bond, V.E. Turner, L.D. Brown, D.J. Bashford, J.E. Cussans, and R.A. Nicholson. 2010. Excavations at Old Scatness, Shetland Volume 1: The Pictish Village and Viking Settlement. Shetland Heritage Publications, Lerwick, Shetland. Dugmore, A.J., C. Keller, and T.H. McGovern. 2007. Norse Greenland settlement: Reflections on climate change, trade, and the contrasting fates of human settlements in the North Atlantic islands. Arctic Anthropology 44(1):12–36. Fenton, A. 1978. The Northern Isles: Orkney and Shetland. Donald, Edinburgh, UK. Forster, A.K. 2004. Shetland and the trade of steatite goods in the North Atlantic region during the Viking and Early Medieval period. Ph.D. Thesis. University of Bradford, UK. Forster, A. 2005. Steatite, resource control, and the Orkney earldom. Pp. 55–74, In O. Owen (Ed.). 2005. The World of Orkneyinga Saga. The Orcadian Limited, Orkney. Forster, A.K. 2007. The early Viking steatite assemblage. Pp. 20–51, In B. Ballin Smith (Ed.). Norwick, Unst. Shetland: An Excavation of Early Viking Settlement. Shetland Amenity Trust, Lerwick, UK. Forster, A.K., and J.M. Bond. 2004. North Atlantic networks: Preliminary research into the trade of steatite in the Viking and Norse periods. Pp. 218–229, In R. Housley and G. Coles (Eds.). 2004. Atlantic Connections, Adaptations: Economies, Environments, and Subsistence in the North Atlantic Realm. AEA/NABO Conference proceedings, University of Glasgow 2001. Oxbow Books, Oxford, UK. Forte, A., R. Oram, and F. Pedersen. 2005. Viking Empires. Cambridge University Press, Cambridge, UK. Friedland, K. 1983. Hanseatic merchants and their trade with Shetland. Pp. 1–28, In D.J. Withrington (Ed.). 1983. Shetland and the Outside World 1469–1969. Oxford University Press,Oxford, UK. Gaimster, M. 1992. Scandinavian gold bracelets in Britain. Money and media in the Dark Ages. Medieval Archaeology 34:1–28. Graham-Campbell, J., and C.E. Batey. 1998. Vikings in Scotland. An Archaeological Survey. Edinburgh University Press, Edinburgh, UK. Hamilton, J.R.C. 1956. Excavations at Jarlshof, Shetland. Her Majesty’s Stationery Office, Edinburgh, UK. Hansen, G. 2015. Itinerant craftspeople in 12th-century Bergen, Norway. Pp. 28–50. In G. Hansen, S. P. Ashby, and I. Baug. (Eds.) Everyday Products in the Middle Ages: Crafts, Consumption, and the Individual in Northern Europe c. AD 800–1600. Oxbow Books, Oxford, UK. Helle, K. 1995. Part I: Down to 1536. Pp. 3–115, In R. Danielsen, S. Dyrvik, T. Grønlie, and K. Helle, E. Hovland (Eds.). Norway: A History from the Vikings to Our Own Times. Scandinavian University Press, Oslo, Norway. Helle, K. (Ed.). 2003. The Cambridge History of Scandinavia. Volume I. Prehistory to 1520. Cambridge University Press, Cambridge, UK. Hunter, J.R. 1994. “Maritime culture”: Notes from the land. The International Journal of Nautical Archaeology 23(4):261–264. Ingimundarsson, J.H. 1992. Spinning goods and tales: Market, subsistence and literary productions. Pp. 217–230, In G. Pálsson (Ed.). 1992. From Sagas to Society: Comparative Approaches to Early Iceland. Hisarlik Press, Enfield Lock, UK. Ingstad, H., and A.S. Ingstad. 2001. The Viking Discovery of America: The Excavation of a Norse Settlement in L’Anse aux Meadows, Newfoundland. Checkmark Books, New York, NY, USA. Jakobsen, J. 1928. An Etymological Dictionary of the Norn Language in Shetland. David Nutt/Vilhelm Prior, London, UK/Copenhagen, Denmark. Jones, R.E., V. Kilikoglou, V. Olive, Y. Bassiakos, R. Ellam, I.S.J. Bray, and D.C.W. Sanderson. 2007. A new protocol for the chemical characterisation of steatite—Two case studies in Europe: The Shetland Islands and Crete. Journal of Archaeological Science 34(4):626–641. Larsen, A.C. 2013. The excavation at Belmont, Wadbister, Unst, Shetland. Pp. 181–214, In V. Turner, J.M. Bond, and A.C. Larsen. (Eds.). 2013. Viking Unst. Shetland Heritage Publications, Lerwick, Shetland. Lidquist, O. 1997. Peasant Fisherman Whaling in the Northeast Atlantic Area, ca. 900–1900 AD. University of Akureyri, Akureyri, Iceland. Lopez, R. S. 1976. The Commercial Revolution of the Middle Ages, 950–1350. Cambridge University Press, Cambridge, UK. Lossius, S.M. 1977. Kleberkarmaterialet fra Borgund, Sunnmøre. Arkeologiske Avhandlinger fra Historisk Museum, Universitet i Bergen, Bergen, Norway. Lossius, S.M. 1979. Klebermaterialet. Pp. 64–71, In E. Schia. (Ed.). De Arkeologiske Utgravinger I Gamlebyen, Oslo. Oslogate 3 og 7, Olso, Norway. Miller, W.I. 1990. Bloodtaking and Peacemaking. Chicago University Press, Chicago, IL, USA. Morris, C.D., C.E. Batey, and D.J. Rackham. 1995. Freswick Links, Caithness: Excavation and Survey of a Norse Settlement. Highland Libraries with the Highland Archaeology Service, Inverness, UK, and the North Atlantic Biocultural Organization, New York, NY, USA. Morrison, I.A. 1973. Jarlshof. The marine environment of a Viking settlement. Nautical Archaeology 2(2):382–384. Mykura, W. 1976. British Regional Geology: Orkney and Shetland. Her Majesty’s Stationary Office, Edinburgh, UK. Owen, O., and C. Lowe. 1999. Kebister. The Four- Thousand-Year-Old Story of One Shetland Township. Society of Antiquaries of Scotland, Edinburgh, UK. Pálsson, H., and P.T. Edwards. (Eds.). 1978. Orkneyinga Saga: The History of the Earls of Orkney. Penguin Books, London, UK. Paterson, C. 1997. The Viking Age trefoil mount from Jarlshof: A reappraisal in the light of two new discoveries. Proceedings of the Society of Antiquaries of Scotland 127:649–657. 2016 Journal of the North Atlantic No. 29 J. Marttila 20 Perdikaris, S. 1996. Scaly heads and tales: Detecting commercialization. Pp. 21–33, In A. Morales (Ed.). Early Fisheries, Archaeofauna. Ichthyoarchaeology and the Archaeological Record. Proceedings of the 8th Meeting of the ICAZ Fish Remains Working Group. Madrid, Spain Perdikaris, S. 1998. The transition to a commercial economy Lofoten fishing in the Middle Ages, a preliminary report. Anthropozoologica 25–26:505–510 Perdikaris, S. 1999. From chiefly provisioning to commercial fishery: Long-term economic change in Arctic Norway. World Archaeology 30(3):388–402 Perdikaris, S., G. Hambrecht, S. Brewington, and T. Mc- Govern 2007. Across the fish event horizon: A comparative approach. Pp. 51–62, In H.H. Plogmann. The Role of Fish in Ancient Time. Verlag Marie Leidorf, Westphalia, Germany. Platt, M.I. 1956. Report on animal bones. Pp. 214–215, In J.R. C.Hamilton (Ed.). Excavations at Jarlshof, Shetland. Her Majesty’s Stationery Office, Edinburgh, UK. Resi, H.G. 1979. Die Specksteinfunde aus Haithabu. Berichte uber die Ausgrabungen in Haithabu. Karl Wachholtz Verlag, Neumünster, Germany. Ritchie, P.R. 1981. Rock as raw material for early Scottish settlers. Pp. 7–14, In J. Kenworthy (Ed.). Early Technology in North Britain. Edinburgh University Press, Edinburgh, UK. Ritchie, P.R. 1984. Soapstone quarrying in Viking lands. Pp. 59–84, In A. Fenton and H. Palsson (Eds.). Northern and Western Isles in the Viking World: Survival, Continuity, and Change. John Donald, Edinburgh, UK. Schou, T.P. 2015. The soapstone vessel production and trade of Agder and its actors. Pp. 204–228, In G. Hansen, S.P. Ashby, and I. Baug (Eds.). Everyday Products in the Middle Ages: Crafts, Consumption, and the Individual in Northern Europe c. AD 800–1600. Oxbow Books, Oxford, UK. Sephton, J. 1994 [1899]. Sverrissaga: The Saga of King Sverri of Norway. Facsimile reprint. Llanerch Publishers, Felinfach, UK. Skjølsvold, A. 1961. Klebersteinindustrien i Vikingetiden. Oslo-Bergen, Norway. Skjølsvold, A. 1969. Et keltertids klebersteinsbrudd i kvikne. Viking 33(1969):201–238. Small, A. 1966. Excavations at Underhoull, Unst, Shetland. Proceedings of the Society of Antiquaries of Scotland XCVIII:225–248. Smith, B. 1984. What is a scattald? Rural communities in Shetland, 1400–1900. Pp. 99–124, In B.E. Crawford (Ed.). Essays in Shetland History. Shetland Times, Lerwick, Shetland, UK. Sognnes, K. 1979. The relationship between coastal and inland areas in the Viking period of West Norway. Acta Archaeologica 50(1979):223–233. Stewart, J. 1987. Shetland Place-Names. Shetland Library and Museum, Lerwick, Shetland. Stummann Hansen, S. 1991. Toftanes: A Faroese Viking Age farmstead from the 9–10th centuries A.D. Acta Archaeologica 61:44–53. Turner, V. 1998. Vikings on the hillside at Catpund, Cunningsburgh. Pp. 95–106, In V. Turner (Ed.). The Shaping of Shetland. The Shetland Times, Lerwick, Shetland, UK Turner, V., S. Dockrill, and J. Bond. 2005. Viking settlement in an Iron Age village: Old Scatness, Shetland. Pp. 245–249, In A. Mortensen and S.V. Arge (Eds.). 2005. Viking and Norse in the North Atlantic: Selected Papers from the Proceedings of the Fourteenth Viking Congress, Tórshavn, 19–30 July 2001. Føroya Fróðskaparfelag (The Faroese Academy of Sciences) in collaboration with Føroya Fornminnissavn (Historical Museum of the Faroe Islands), Tórshavn, Faroe Islands, UK. Turner, V.E., P. Sharman, and S. Carter. 2009. Excavations at Catpund, Cunningsburgh. Pp. 70–84, In A. Forster and V.E. Turner (Eds.) 2009. Kleber: Shetland’s Oldest Industry. Shetland Soapstone since Prehistory. Shetland Amenity Trust, Lerwick, Shetland, UK. Turner, V.E., J.M. Bond, and A.C. Larsen. 2013 Excavation and Survey in Northern Shetland 2006–2010. Viking Unst. Shetland Heritage Publications, Lerwick, Shetland, UK. Urbanczyk, P. 1992. Medieval Arctic Norway. Institute of Archaeology and Ethnology, Warsaw, Poland.