Introduction

The Melkøya project is an interdisciplinary project based on archaeological excavations on the island of Melkøya, Arctic Norway. Large-scale excavations were conducted in 2001 and 2002. The project will be completed in 2006.

Melkøya is located in the strait of Sørøysundet, at 70°, 40 N in Finnmark, Arctic Norway (Figure 1). The island is less than 1 ha with a maximum elevation of 71m asl. With the exception of two cobble beaches on the more sheltered south side of the island, the coastline consists of sheer cliffs of bedrock. Sites were only identified in or around these two beaches, even though the whole island was intensively surveyed and test pitted. Clusters of sites were identified in three areas: Normannsvika, Kilden and Sundfjæra (Figure 1), which chronologically cover the time period from the 'initial pioneer phase' (Reference BjerckBjerck 1995) of the Early Stone Age (10 000 BP) to the Sámi Iron Age (1500 BP).

Figure 1.

Aerial view of Melkøya and find areas, insert of Northern Europe and the location of Melkøya in lower right corner. Figure by Anja Roth Niemi.

Figure 2.

Due to the effects of the Gulf Stream the coastal climate is unusually mild for its latitude, but the weather is unpredictable as illustrated in these pictures which were taken all within the same week. Pictures copyright Tromsø Museum.

The climate is oceanic, the sea never freezes in winter and the annual median temperature is + 2° C (Figure 2). The weather can be highly unpredictable with rough seas, windy conditions and frequent storms. Due to the location far above the Arctic Circle the people also had to cope with dramatic changes in an environment with two summer months of constant daylight, to an equally long period in winter when the sun does not rise above the horizon. For people living by and from the sea, such conditions clearly structured social life as well as resource exploitation, habitation and how the landscape was used.

Traditionally small and exposed islands like Melkøya are viewed as marginal even for Stone Age economies. Nevertheless, the excavations have documented considerable amounts of settlement data from all prehistoric periods. Apart from the successful cultural adaptation to these conditions one has to take into consideration that Melkøya is located in the middle of a strong tidal current channel. Such channels have a particularly high plankton production which attracts large numbers of fish, marine birds and sea mammals making them one of the most optimal coastal biotopes for marine exploitation (Reference OlsenOlsen 1998, Reference BergsvikBergsvik 2001: 4).

Aims

The majority of Stone Age sites in Norway are defined by the presence of stone artefacts or semi-subterranean houses. Excavations have traditionally focused on the distribution of artefacts or on the house structures themselves (Glørstad 2002: 20, Reference GrønGrøn 2000: 188-189, Reference RamstadRamstad 2003). This bias has not only reduced the amount of alternative data but also limits our understanding of past settlements (Grøn 2003: 218-219). The main goal of the project was to map, excavate and identify the function of as many of the prehistoric structures and associated features on the island as possible. By giving equal attention to multiple sets of data we hope to stimulate new and alternative perspectives on the organization and structuring of space at hunter-fisher settlements.

Figure 3.

To get a more viable picture of hunter-fishers' impact, as well as natural changes, on the local environment extensive palaeobotanical investigations were undertaken (Reference JensenJensen 2004). Analyses of 5 peat cores combined with a number of in situ samples has created a more detailed picture of local vegetation types and anthropogenic impact extending back almost 10 000 years ago (Fig from Jensen fig. 12, 2004:282).

Figure 4.

The relationship between structures in Normannsvika and the spatial variations of phosphates and magnetic susceptibility (Figure after Johan Linderholm fig. 1 and 2, 2003:49).

Results

Mechanical surface stripping was used in order to explore as much as possible of the total extent of the past settlement areas. During the first season, topsoil was removed from large areas in and around the sites at Kilden and Normannsvika, while during the work in Sundfjæra in 2002 we were able to expose all suitable settlement areas.

A large number of structures were identified from semi-subterranean houses and lighter dwelling types and tent rings, to a variety of hearths, cooking pits, storage pits, activity areas, dumping zones, middens and grave remains (Reference RamstadRamstad 2003 a; b). These traditional archaeological investigations were combined with extensive palaeobotanical investigations (Figure 3) (Reference JensenJensen 2004), chemical soil analyses (Figure 4) (Reference LinderholmLinderholm 2003) and systematic investigations of traditionally 'unpromising' parts the settlement areas (Figure 5) (Reference RamstadRamstad 2003). The results clearly indicate that hunter-fisher settlements consisted of more than just dwellings and areas for tool production. This diverse data also raises important questions regarding traditional concepts of settlement.

Figure 5.

Among finds recovered in the traditionally 'neglected' areas, were 8 amber beads found in a small crevice in the bedrock some distance away from the nearest house site. The amber is identified as originating from the Baltic region proving the existence of networks connecting Melkøya to areas several thousand kilometres away (Reference RamstadRamstad 2004). Picture by Adnan Icagic, copyright Tromsø Museum.

Figure 6.

A topographic map of Sundfjæra showing the relation between sites and prehistoric shorelines, this is illustrated by the fact that the oldest sites are at the highest elevation and the younger ones are closer to the present-day sea level. The close spatial relations between landscape, topography and structures are further investigated on the basis on micro-topographical maps. Illustration by Anja Roth Niemi.

Both Kilden and Normannsvika have provided new data for the study of settlement patterns and intra-site relationships during the Early Metal period and Late Stone Age, respectively. From the upper to the lower Sundfjæra, all prehistoric periods from the beginning of the Early Stone Age to the Sámi Iron Age were documented (Figure 6). Treating Sundfjæra as a single settlement unit gives a unique diachronic picture of the way the landscape was utilized through time, thus permitting studies of stability and changes in the structuring of space over considerable time periods.

Conclusion Due to its privileged location in a landscape the seasonal and spatial distribution of resources must have allowed for a high variety of choices for the people living at Melkøya. The chronological variation of settlement data collected through the project highlights how the available food resources structured or constrained past settlements on the island, but did not determine them. The rich and varied material demonstrates how the island always has been involved in networks stretching over large parts of Northern-Scandinavia. Thus small islands can no longer be viewed as either marginal or isolated places for Stone Age societies. The results from the project have generated new perspectives giving a unique and detailed picture of the use and re-use of one island through long periods of time and the ability to provide a complete picture of spatial and temporal transformations over entire settlement sequences.

Further information can be found on the project's home page at http://www.uit.no/melkoya/.