Copper production in Bronze Age Norway? CHRISTOPHER PRESCOTT This article deals with the question of whether or not there was non-ferrous metal extraction on the Scandinavian Peninsula during the Late Neolithic and Bronze Age. This initial enquiry refers empirically to a case study from Årdal in western Norway. Before describing this case, some general research historical premises are examined – to suggest some of the ontological reasons for the present state of this ield in archaeometallurgy. The introductory comments also provide background for a discussion of the consequences the future documentation of early copper production in Scandinavian might have: The acceptance of prehistoric exploitation of Scandinavian ores must certainly have an ontological and interpretative impact on our views of the Bronze Age. The three-period system, encompassing the Stone Age, the Bronze Age and the Iron Age, is one of the major achievements of Scandinavian archaeology. Formulated by Christian J. Thomsen in the early 1800’s (Klindt-Jensen 1975:50f, Trigger 1989:75f), and published in 1836, twenty years before Darwin’s (1968[1859]) The Origin of Species, Thomsen’s deceptively simple evolutionary system has been elaborated, criticised and rediscovered by generations of archaeologists. It remains fundamental to what we teach, how we express our selves, indeed how we think about the Bronze Age. In the history of Scandinavian Bronze Age archaeology another milestone occurred ifty years later, when Montelius divided the Bronze Age into six periods (Montelius 1885, 1903). Montelius’ system has stood the test of time, and remains the referential standard for Nordic Bronze Age chronology, especially for period-internal divisions. This system is based on the application of the premise of continent-wide diffusion to deine the chronological pegs for the start and end of the Bronze Age. 130 years later the legitimacy of analysing the dialectic interaction between continental trends with regional and local structures and traditions remains a cornerstone of Bronze Age research in some contemporary Scandinavian archaeology (Larsson 1997, Johansen 2000, Kristiansen 1998, Prescott & Walderhaug 1995, Weiler 1994, Østmo 1997). Despite the undisputable value of a continuously developing 200 year long tradition of research based on Thomsen and Montelius, certain inherent premises have perhaps developed in a virtually dogmatic direction. One such tenet holds that all metal was imported to Scandinavia, and was not locally mined until the introduction of ferrous technology. The following article does not offer conclusive evidence of copper production from sources on the Scandinavian Peninsula – it offers circumstantial support for a more open attitude to the question (cf. Stenvik 1988; også Johansen 2000:26ff, Prescott 1999:218). Indeed, it is more reasonable to assume that there was some level of copper production, than to assume that there was not. The Nordic Bronze Age, which developed its own vibrant expression at the beginning of the Early Bronze Age period II, is generally regarded as one of the richest expressions of European Bronze Age metal working (Alexander 1980:222). Therein lays an often mentioned paradox; all the metal utilised in this prehistoric metallurgical blossoming was assumed to have been imported from central or Western Europe. Though copper objects were imported to Scandinavia in the Funnel Beaker Neolithic (TRB), and indeed some exploitation of Northern European sources of copper might have started at this 183 time (Klassen 2000: 211ff), more vigorous metallurgical inluences are seriously manifested as of the Late Neolithic I (LNI). Still, the early TRB expressions of metallurgy are interesting within the context of the present article, as they suggest that in addition to the external Bronze Age inluences, a regional knowledge of metallurgy might have been established before the LN/BA. The earliest larger scale indications of metalworking in Scandinavia are found in the Late Neolithic II (LNII), and are primarily casting of langed axes based on imported metal similar to that used in Dutch Bell Beaker and Danubian Bronze Age Cultures (Vandkilde 1996). BRONZE AND THE NORWEGIAN BRONZE AGE Direct evidence of local casting is found in the moulds, crucibles and metal residues that occur on settlement sites, in association with burial grounds, and as inds outside of a primary context. The end products of the casting process – the metal artefacts – are primarily found in graves and hoards – not on settlements. Within the borders of present day Norway there are relatively few bronzes – slightly more than 800. However, the quality of the bronzes, evidence of melting and casting, as well as the overall cultural expression (cf. Shetelig 1925) suggests that bronze was an integrated element of the material culture probably from around the LNII/EBAI transition (Prescott 2000, Johansen 2000). The low number of bronzes recovered in present-day Norway as compared to more central Nordic regions might conceivably be a result of the nature of demands generated by the social context – i.e., less hierarchical social organisations with less demand for prestigious displays of metal compared to societies in southern Scandinavia. The meagre number of bronzes might also be source-critically associated with the most common form for burial known in most parts of southern and central Norway – the rock cairns, as such depositional environments provide unfavourable conditions for the preservation of bronzes. A number of other source critical arguments can also be offered (ref. Bakka 1980, Brøgger 1925:105f, Johansen 1981;114ff, Prescott 2000, Shetelig 1925:91).Still, it has usually been implied that little bronze found its way to the northern peripheries outside of southern 184 Scandinavia. On a somewhat provocative note, could it be argued that there is little direct correlation between the amount of archaeologically recovered metal objects in Norway and the amount or importance of metal used in the Bronze Age, and that the lack of recovered metal objects is a function of social conditions in prehistory, depositional practices and post-depositional factors? Could some regions in Scandinavia, also those with few recovered bronzes, actually have been net exporters of copper – not importers? EARLY METALLURGY IN NORWAY The archaeological evidence of metallurgy (crucibles, moulds, metal residues) suggest that the casting of copper alloys was far more widespread than the bronzes recovered from depots and graves suggests (cf. Oldeberg 1943;134, Bakka 1976, Solberg 1988). As Egil Bakka (1976:16) and Øystein Johansen (1986:77) observed sometime ago, the moulds and crucibles have a distribution far beyond – indeed almost at odds with – what archaeology has traditionally viewed as the most central localities for Bronze Age communities around the Oslo fjord, in South-western Norway and in Trøndelag. It is not only the geography that is «odd» in this respect: for although crucibles and moulds are recovered in connection with long houses (cf. Løken 1987:239) in Rogaland or cemeteries in Østfold (Aninset, Melheim & Prescott in press, Wangen 1998), such inds are also found along the coast, in the inland and in the highlands. They occur at «open settlement sites» and in rock shelters with an artefactual inventory otherwise characterised by bifacial lithics and bone tools (i.a., Prescott 2000). Thus, although the most striking metal tools, weapons and adornments primarily stem from areas with concentrations of metal inds in or near landscapes with the greatest agricultural potential, traces of casting –moulds, crucibles, metal residues – are found on habitation sites in a much greater region. Traces of metal in such contexts are often dificult to identify, and consist of small, broken pieces recovered from hearths and «cultural layers» (i.e. refuse deposits). We must assume that these small bits and pieces were unintentionally lost, i.e. a depositional history completely different than that of the depots and graves that are the basis for traditional studies. The recovery of these small, sometimes microscopic, pieces is strongly dependent on the method of excavation. Trowel excavation of large areas, combined with systematic sieving with relatively ine meshes, especially of hearth deposits, is conducive for the retrieval of such evidence. The remains of metal and the traces of small-scale casting suggest that copper and bronze probably were more common than a face-value interpretation of the archaeological record indicates. EXPLOITABLE ORES IN SCANDINAVIA? Are there local sources of copper in Norway that could have been exploited with Bronze Age technology? There are many deposits of pure copper, copper-bearing rock like malachite, oxides, occurrences in conjunction with greenstones, quartzite, and sulphide deposits, as well more or less pure copper in association with hydrothermal deposition or secondary weathering and recrystallization (Brünnich 1777:26f, Neumann 1985: 2f, Oftedahl 1980:109). A number of these have been mined in the historical times, the silver mines at Kongsberg and the copper mines at Røros being the most well-known examples. There are also numerous small-scale mines, for example in Kvitseid, Dalane, Porsanger, Berg and Årdal. Similar occurrences are known from Sweden, for example at Stekenjokk in Sweden (Neumann op. cit.). Two important factors should be summed up here. First, small scale, surface occurrences are widespread, – and not systematically charted. Second, the ore qualities are variable, but numerous, – also uncharted – occurrences may be readily exploited without resorting to the complicated processing of pyretic ores. In the Bronze Age one could easily imagine unobtrusive exploitation of small surface occurrences, and using local information (as opposed to large scale mapping of ore sources) to identify small-scale sources might be more proitable than pursuing sources that could have been exploited on an industrial scale. As far as the present author knows, there has never been a systematic attempt by researchers in the ield to see if prehistoric exploitation of small-scale sources might have occurred. From a sourcecritical point of view, one would expect the most readily identiiable sources to have been exploited both in prehistoric and historic times. Early smallscale traces would therefore often be disturbed or obliterated by later mining activities (Craddock 1995). Local accounts of mining endeavours can therefore provide interesting ethno-historical data from low-level technological extraction and processing of ores, and thereby provide analogies for research of prehistoric metallurgy – and clues to where copper sources might be found. The Bronze Age population was part of a network that placed a premium on metals. There are sources of copper on the Scandinavian Peninsula, and though of variable quality, some of these could have been readily identiied in prehistory, as the example below should demonstrate. The traces of Bronze Age melting and casting, the moulds, crucibles and metal residues, demonstrate that signiicant elements of the metallurgical process were mastered as of the LN-EBA transition. The whole cultural expression of the Late Neolithic II/Bronze Age demonstrates that large-scale interaction was intense, and that comprehensive systems of knowledge were rapidly transferred over large parts of the Old World, also to Scandinavia (cf. Larsson 1997, Vandkilde 1996: 294ff). Are there reasons to assume that this would not also pertain to knowledge regarding the exploitation of copper occurrences? The quarrying of soapstone for moulds demonstrates that local resources were mined in conjunction with metallurgy during the Nordic Bronze Age. Several studies demonstrate that people in the Bronze Age were deeply knowledgeable of the totality of the landscape. They ploughed the ields, herded the locks, boated, hunted, ished and trapped, as well as quarried stone for tools, soapstone for moulds and vessels and clay for pottery. Would not these people have a detailed knowledge of the natural environment and its mineralogical resources? We can thus assume that the Bronze Age population of the Scandinavian Peninsula had the technological knowledge to exploit copper, that people were familiar with the mineral resources in their environments, and that they were used to quarrying. In general terms it therefore seems more likely that local sources were exploited, than that they were not exploited. Below a speciic case from Årdal in Sogn, western Norway is presented. The circumstantial evidence begs a campaign to systematically investigate the possibility of Scandinavian copper production in the Bronze Age. The potential 185 for exploitation of local metal sources has been suggested in connection with the Årdal studies in previous publications; however the present article attempts to evaluate the case more systematically. THE CASE FROM ÅRDAL IN SOGN, WESTERN NORWAY Some years ago the author was involved in ield research along the Nyset-Steggje drainage systems (Bjørgo, Kristoffersen & Prescott 1992) and in the Skrivarhelleren rock shelter (Prescott 1992) in the highlands of Årdal, in Sogn, western Norway (ig.1). An interesting aspect from these sites is the traces of metallurgy (cf. Prescott 2000, Johansen 2000). These traces are intriguing because they come from highland environments not usually thought of as Bronze Age landscapes, 790-1100 metres above sea level (m asl), exploited seasonally for pastures and wildlife resources, and because the region has otherwise only yielded a handful of bronzes. The recovery of the inds from Nyset-Steggje and Skrivarhelleren is partially a result of the excavation method. Most sediment was excavated by hand and sieved though ine meshes (2&4 mm), and some «ochre» samples from hearths were subjected to microscopic analysis. The metal and metal-related inds from Late Neolithic and Bronze Age contexts are summed up in table 1. The soapstone mould, the chopped piece of bronze (probably for re-melting) and the fragmentary crucible indicate that even on these seasonally exploited, highland sites – far away from the centres of Bronze Age culture, – people were familiar with and for some reason cast bronze. The microscopic traces of bronze were found on the edge of a hearth in a sample of ochre. Glazing on the grain surface indicates vitriication of the silicates in the samples, which again indicates temperatures conducive for casting bronze. The analyses were carried out by Kirsti Riisøen at the University of Bergen and she concludes that the samples had been subjected to temperatures of around 1100–1200 Celsius – in good accordance with the melting point for bronze. In short this indicates that the casting was carried out at hearths at the Skrivarhelleren rock shelter 790 m asl. Suficient temperatures must have been achieved at the open hearth through the use of bellows. Fi186 nally, the date of the context should be noted: the LNII. The inds from Nyset-Steggje and Skrivarhelleren are not unique, but they are especially fascinating because they were assembled as part of a systematic research campaign through a period of nearly ten years. They are also important because of the unusual depositional, environmental and geographic contexts that challenge our concepts of the relationship between Bronze Age society and culture. The initiation of metallurgy in the late LN/early EBA, and its continuous practice through the Bronze Age is also remarkable. The chopped piece of bronze from layer V (EBA) indicates that inished objects were re-melted and re-cast. But was all the copper necessarily imported – or could local copper sources have been exploited? LOCAL COPPER EXTRACTION? It is interesting to note that in the AD 1700’s there were copper mines in the valleys and mountains across the fjord from the Nyset-Steggje and Skrivarhelleren sites, i.e. just north of the innermost part of the Sognefjord (ig. 1). Sources of copper were found around the valleys of Nundalen and Fardalen, and in the mountains above called Gruvefjellet («Mine Mountain») and Grøndalsfjellet. These mines are about a day’s journey by foot and boat from Skrivarhelleren. Indeed, one registered copper claim, never exploited, was found at Brennborg a few hours walk from the Steggje valleys, on the same side of the fjord as the Nyset and Steggje river systems. The botanist dr. philos. Søren Ve wrote an extensive account of the copper mines in Årdal (Ve 1971a &b). The mines were primarily exploited in the 1700’s, but Ve’s accounts of how the mines were found, how the miners regarded the occurrences of copper and of the geology are of interest here. Ve described the initial discovery of ore in Årdal: «One day in June 1700, two men were hunting reindeer in the mountains between Fardalen and Seimsdalen. They were Botolv Endersen Hæreid and Kristoffer Ingebriktsson Årebru. After a while they came all the way up to Grøndalsfjellet, 1500 m asl. Here, Botolv noticed some rocks that were dif- Figure 1. Map of Årdal. Archaeological sites and placenames referred to in the text are indicated. Map: Anne Engesveen. ferent from those otherwise found in the ground around here, for example in regard to their colour and weight. He pointed this out for Kristoffer, and said it was ore. He put a few pieces in his sack, and walked on. It may seem strange that a farmer might appreciate something like this. However, Botolv lived in a time of great interest for ores and mining…» (Ve 1971b:384, all translations by Prescott)1One could easily imagine this scenario taking place thousands of years earlier - during the Late Neolithic or Bronze Age. The next deposit of copper-bearing rock was found the following year at lower altitudes (847 m asl) and closer to the fjord, at Blåberg («Blue Mountain»). It was discovered by Overkonduktör (technician) Lorentz Schioldal from Bergen who visited Årdal in 1701, in the wake of the discovery of copper the year before. The account of his discovery is no less surprising than that of the previous year: [From Årdal]«…he noticed the blue-green colour at Blåberg between Nundalen and Fardalen. He thought that the hue might be caused by copper rust –“weathering” he called it – and that this must be copper ore. He therefore went up to Blåberg looking for ore, and found two ore-bearing dykes, “veins”.» (Ve 1971b:387)2 The miners that came to work in the mines that were developed after the irst discoveries of ore in Årdal had this impression of the surface occurrences: 187 Finds Date** Skrivarhelleren rock shelter, layer VI Find context Small piece of copper alloy*, fnr 189 LNII Prescott 1992:79 Skrivarhelleren rock shelter, layer VI Rivet head, copper alloy*, fnr187 LNII Prescott 1992:79 Skrivarhelleren rock shelter, layer VI Microscopic fragments of bronze, found in ochre from edge of a hearth, fnr. 948 & 949 LNII Prescott 1992:78f, 81f Skrivarhelleren rock shelter, layer V Massive piece of copper alloy* with chopping marks, fnr 190 EBAI/II - LNII Skrivarhelleren rock shelter, layer VI Wedge-shaped pendant; (touchstone?), fnr. 447/448 Skrivarhelleren rock shelter, layer IV Fragment of a soapstone lanzetten mould, fnr 1 Urutlekråi, hearth, settlement Part of bronze awl LBA/PRIA Bjørgo et al. 1992:116 Kalvebeitet, settlement Fragment of crucible LBA/PRIA Bjørgo et al. 1992:104 LNII LBA V Reference Prescott 1992:78 Prescott 1992:58ff Prescott 1992:76 Table I. Finds indicating non-ferrous metallurgy at Nyset-Steggje and Skrivarhelleren in Årdal, Sogn. * Only visual characterisation, originally catalogued as bronze. ** LN = Late Neolithic, EBA = Early Bronze Age, LBA = Late Bronze Age, PRIA = Pre-Roman Iron Age. «In a review from 1720, Jens Worm writes for example that the blended surface ores from Grønfjellet were rich and promising at the beginning. Ore that is so rich and easy to mine was jokingly called “farmer’s ore” by the miners.» (Ve 1971b:404).3 Based on «old records concerning Årdal’s Copper Works», Ve provides an overview of the potential sources of copper, largely assessable from the surface. These include mixed copper ores, malachite, copper oxide, pure copper, copper pyrites, pyrrhotine, pyrite, galena and magnetite (Ve 1971a:32, comp. Brünnich 1777:26f). Finally, Ve points out that there were very minor occurrences of silver and gold. The occurrences are all associated with a large intrusion of gabbros. CONCLUDING IDEAS This article has demonstrated that there was early copper-based metallurgy – melting and casting – during the Bronze Age and probably already in the Late Neolithic II in Sogn, Norway – as there was a number of other places in Scandinavia. It is also demonstrated, based on ethno-historical studies and mine records, that there were local sources of readily extractable copper in Årdal, and that several of these would have been accessible to and could have been identiied by the local Bronze Age inhabitants. Årdal is but one of many 188 places on the Scandinavian Peninsula where such a group of variables can be expected to be identiied (e.g., Brünnich 1777, Johansen 2000:26f, Stenvik 1988:299, cf. Oftedahl 1980, Neumann 1985). What makes Årdal special, apart from its peripheral status in terms of our understanding of the Bronze Age, is the intensity and scope of archaeological research that was conducted there in the 1980’s. This article does not prove that copper sources were exploited in the Bronze Age. On the other hand, research into whether or not sources of copper on the Scandinavian Peninsula were exploited during the Bronze Age is very limited. This article suggests that it is reasonable to assume some level of copper extraction, and that a programme to seriously investigate the matter should therefore be initiated. If future research was to unequivocally document the exploitation of copper sources in Norway, what would the consequences be? This of course would depend on the scale and nature of such production. However, it would potentially lay the basis for a broader anthropological and ritual analysis of the role of early metallurgy in Scandinavia. Such results would challenge a number of the implicit and explicit premises for the interpretation of the Bronze Age in Norway and Scandinavia. It would also have consequences for the historical understanding of the Bronze Age in Northern Europe. Could such results have consequences for ideas that have underpinned 200 years of Bronze Age research? Acknowledgements I have had the good fortune of working with early metallurgy in collaboration with Lene Melheim and Nils Aninset at Hunn in Østfold. Lene Melheim’s and Claire Glenton’s suggestions based on an early draft of this article are much appreciated. Joakim Goldhahn’s enthusiastic appreciation of the subject spurred a renewed interest in the potential exploitation of Scandinavian copper sources. Christopher Prescott, Institutt for Arkeologi, Konservering og Historiske studier: Universitetet i Oslo. E-mail: christopher.prescott@iakh.uio.no SAMMENDRAG Kobberutvinning i Norge i bronsealderen? Sporene etter bronsestøping (digler, støpeformer og annet støpeavfall) i Norge innes over et stort geograisk område og i mange ulike natur- og kulturmiljøer. Siden Montelius’ tid har det vært vanlig å hevde at grunnlaget for denne produksjonen – kobber og tinn – var utelukkende importert. Det er likevel en rekke faktorer som tyder på at ikke-ferriske malmressurser kan ha vært utnyttet allerede i bronsealderen. Med utgangspunkt i materiale fra Årdal i Sogn, sammenstiller artikkelen noen gruvehistoriske, malmgeologiske og arkeologiske data og hevder at det er rimelig å gå ut fra at man har utvunnet kobber i norsk bronsealderen. NOTES 1 2 Ein juni dag året 1700 gjekk to mann på reinsjakt i fjellet mellom Fardalen og Seimsdalen. Det var Botolv Endreson Hæreid og Kristoffer Ingebriktson Årebru. Til slutt kom dei helt opp på Grøndalsfjellet, godt og vel 1500 m o.h. Der la Botolv merke til noko stein som skilde seg ut frå steinen elles i fjellet der ikring mellom anna i farge og tyngd. Han peika på dette for Kristoffer, og sa at det var erts. Så tok han nokre stykke i skreppa si og gjekk vidare. Det kan synast rart at ein vanleg bonde hadde sans for slikt. Men Botolv levde i ei tid med stor interesse for malm og bergverk. …la han merke til den blå-grøne fargen I Blåberg mellom Nundalen og Fardalen. Han meinte at leten 3 kom av eir – «vitring» kalla han det – og at det matte vera koparerts i fjellet. Difor gjekk han til Blåberg, og leita etter erts og fann to malmgangar, «årar». I eit oversyn frå 1720 skriv Jens Worm t.d.at malmen av brokut kopar i overlata på Grøndalsfjellet til å byrja med var rik og lovande. Malm som var så rik og lett å bryta, vart av gruvearbeidarne spøkefullt kalla «bondeerts». LITTERATURE Alexander, J. 1980: First-millenium Europe before the Romans. In Sherratt (ed.), The Cambridge Encyclopedia of Archaeology. Cambridge University Press, Cambridge: 222-226. Aninset, N., A. L. Melheim, C. Prescott in press: Hunnundersøkelsen: Landskap, kosmologi, riter & metallurgi. Rapport fra Nordisk Bronsealder Symposie 2000. Copenhagen. (also http://www.hf.uio.no/iakk/hunn/artikkel%20 bronsealdersymposium.htm). Bakka, E. 1976: Arktisk og nordisk i bronsealderen i NordSkandinavia. Miscellanea 25, Trondheim. Bakka, E. 1980: Representativitetsproblem i vestnorsk bronsealder. In Thrane (ed.), Broncealderbebyggelse i Norden. Beretning fra det andet nordiske symposium for broncealderforskning. Skrifter fra Historisk Institut, Odense Universitet, nr. 28, 37-58. Odense. Bjørgo, T., S. Kristoffersen & C. Prescott 1992: Arkeologiske undersøkelser i Nyset-Steggjevassdrgaene 1981-87. Arkeologiske Rapporter16. Universitetet i Bergen. Brünnich, M. T. 1777: Forsøg på Mineralogie for Norge. Det Kongl. Norske VidenskabersSelskab, Tronheim. Opptrykk 1977, NTH, Trondheim. Brøgger, A. W. 1925: Det norske folk i oldtiden. Instituttet for sammenlignende kulturforskning, VIa. Oslo. Craddock, P. T. 1995: Early Metal Mining and Production. Edinburgh University Press, Edinburgh. Darwin, C. 1968 [1859]: The origin of species by means of natural selection or the preservation of favoured races in the struggle for life. Penguin Classics. Johansen, Ø. 1981: Metallfunnene i østnorsk bronsealder. Kulturtilknytning og forutsetninger for en marginalekspansjon. Universitetets Oldsaksamlings Skrifter, Ny rekke 4. Oslo. Johansen, Ø. 1986: Tidlig metallkultur i Agder. Universitetets Oldsaksamlings Skrifter, Ny rekke 8. Oslo. Johansen, Ø. K. 2000: Bronse og makt. Andresen & Butenschøn, Oslo. Klassen, L. 2000: Frühes Kupfer im Norden. Untersuchungen zu Chronolgie, Herkunft und bedeutung der Kupferfunde 189 der Nordgruppe der Trichterbecherkultur. Jysk Arkeologisk Selskap, Moesgård. Hagen. Arkeologiske Skrifter Historisk Museum 4-88, 277-291. Bergen. Klindt-Jensen, O. 1975: A History of Scandinavian Archaeology. Thames and Hudson, London. Stenvik, L. 1988: Steinkøller med skaftfure. In Indrelid, Kaland & Solberg (eds.), Festskrift til Anders Hagen. Arkeologiske Skrifter Historisk Museum 4-88, 292-300. Bergen. Kristiansen, K. 1998: Europe Before History. Cambridge University Press, Cambridge. Larsson, T.B. 1997: Materiell kultur och religiösa symboler. Arkeologiska studier vid Umeå Universitet 4. Umeå. Løken, T. 1987: Forsand: Nå også med den første kjente bronsealderlandsby i Norge. Fra haug ok heidni 4/12. Arkeologisk museum i Stavanger. Montelius, O. 1885. Om tidsbestämning inom bronsåldern med särskildt afseende på Skandinavien. Kongl. vitterhets historie och antiqvitets akademiens handlingar ; 30. Stockholm. Montelius, O. 1903. Die ælteren Kulturperioden in Orient und in Europa I. Die methode. Stockholm. Neumann, H. 1985: Norges mineraler. NGU-Norges geologiske undersøkelse. Skrifter 68. Universitetsforlaget, Trondheim. Oldeberg, A. 1943: Metallteknikk under förhistorisk tid, del II. Lund. Oftedahl, C. 1980: Geology of Norway. NGU-Norges geologiske undersøkelse 356 Bulletin 54. Universitetsforlaget, Trondheim. Prescott, C. 1992: Kulturhistorisk undersøkelser i Skrivarhelleren. Arkeologiske rapporter 14. Universitetet i Bergen. Prescott, C. 1999: Long-term patterns of non-agrarian exploitation in southern Norwegian highlands. In Fabech & Ringtvedt (eds.), Settlement and Landscape. Jutland Archaeological Society, Moesgård. 213-224. Prescott, C. 2000: Symbolic Metallurgy – Assessing Early Metallurgical Processes in a Periphery. In Olausson & Vandkilde (eds.), Form, Function and Context. Material culture studies in Scandinavian archaeology. Acta Archaeologica Lundensia series in 8, no. 31, Lund. 213226. Prescott & Walderhaug 1995: The Last Frontier? Processes of Indo-Europeanization in Northern Europe: The Norwegian case. Journal of Indo-European Studies 23/34, 257-381. Shetelig, H. 1925: Norges Forhistorie. Problemer og resultater i norsk arkæologi. Instituttet for sammenlignende kulturforskning, Oslo. Solberg, B. 1988. Steinøkser med skaftfure fra Syd-Norge. In Indrelid, Kaland & Solberg (eds.), Festskrift til Anders 190 Thomsen, C. J. 1836: Ledetraad til Nordisk Oldkyndighed. Det kongelige Nordiske Oldskrift-Selskab, Copenhagen. Trigger, B. 1989: A history of archaeological thought. Cambridge University Press, Cambridge. Vandkilde, H. 1996: From Stone to Bronze. The Metalwork of the Late Neolithic and Earliest Bronze Age in Denmark. Jutland Archaeological Society Publications XXXII, Aarhus. Ve, S. 1971a: Slik vart bygda til. In Ve (ed.), Bygdebok for Årdal. Natur og næringsliv. Band I, Årdal Sogelag, Årdal. 26-77. Ve, S. 1971b: Koparverket. In Ve (ed.), Bygdebok for Årdal. Natur og næringsliv. Band I, Årdal Sogelag, Årdal. 384496. Wangen, V. 1998: Gravfeltet på Gunnarstorp. Et monument over dødsriter og kultutøvelse i 1200 år. In Østmo (ed.), Fra Østfolds oldtid. Fordrag ved 25-årsjubileet for Universitetets arkeologiske stasjon Isegran. Universitetets Oldsaksamlings Skrifter. Ny rekke 25, Oslo. 153-172. Weiler, E. 1994: Innovationsmiljöer i bronsålderens samhälle och idévaårld. Studia Archaeologica Universitatis Umensis 5. Umeå. Østmo, E. 1997: Horses, Indo-Europeans and the Importance of Ships. Journal of Indo-European Studies 25/3-4, 285326.