Stokkebåten fra Siljan

The International Journal of Nautical Archaeology (2008) 37.1: 3–16 doi: 10.1111/j.1095-9270.2007.00159.x

Boats for Rivers and Mountains: Sources for New Narratives about River Travel?

Pål Nymoen

Norwegian Maritime Museum, Bygdøynesveien 37, 0286 Oslo

What is a boat? Our notions of what constitutes a good boat are tested when it comes to logboats. They are often considered to be of low status, both technologically and functionally. However, logboats are more interesting than that. This article presents three logboats dated to the Roman era and late Viking age, found in the region of Telemark, Norway. The principal concern is to discuss how notions about prehistoric boats can be made known, challenged and discussed by examining the three boats in the context of the rivers and lakes in which they were most probably used.

SPB.OlaNUckYRwMCeElOl SPEuFNbOliRshNinEg WLtdNARRATIVES ABOUT RIVER TRAVEL?

Key words: Norwegian river systems, logboats, Roman, Viking.

Boats and ships from coastal regions play important roles in archaeological narratives about the lines of development in the

cultural history of Norway. Interpretations of rock carvings from the Stone and Bronze Ages, as well as research on Viking ships and medieval seafaring, emphasize the central role of boats in cultural history (for example Nielsen, 1905; Christensen, 1982; Christensen, 1985; Kvalø, 2004; Østmo, 2005). The role of the Nordic tradition of clinker con- struction in promoting mobility and overseas contact and trade in coastal areas is especially in focus.

A glance at a map of Norway reveals a unique topography: a rugged coastline cut by deep fjords and long river systems. These connect the coast and the interior, both as hindrances to transit and as ‘highways’ for the transport of goods and people across long distances. Historically, both logboats and other types of small boats have been important in the material culture associated with lakes and rivers. However, compared to the study of coastal boats, research on freshwater vessels is far less common (Hougen, 1941; Klepp, 1984; Christensen, 1996; Christensen, 2000).

In this article, I will present three logboats found in the region of Telemark in Norway. The contexts of the finds are similar, and the boats share certain formal features. I will describe and document these boats by means of technical drawings. My principle concern, however, is to discuss how notions about prehistoric boats can be made known, challenged and discussed by examining the three boats in the context of the rivers and lakes in which they were most probably used. Further, I would like to discuss what these boats can tell us about the representativeness of the archaeological sources with particular regard to our knowledge about the use of inland waters in coastal areas.

Logboats, in common with boats that are generally referred to as ‘inland boats’ in Norway, have been treated in a rather slighting manner both by cultural heritage and preservation institutions in Norway and by scholars. Two of the boats that I examine in this paper have been kept in storage on private farms for the last 45 years, and the third has been kept in a storage building without a regulated climate at the local museum for the Telemark region. This kind of find situation appears to be typical for logboats, so the possibility of discovering more logboats in storage would seem to be quite good.

Three logboats, two river systems

In 1930 a boat was found at the edge of a bog near a small lake about 25 km from the coast at Siljan, Telemark, Norway. A narrow river winds around the north and south ends of the lake. The lake itself lies in the middle of a river system which connects a broad, mountainous region in the north with a region of lowlands, large lakes

© 2007 The Author. Journal Compilation © 2007 The Nautical Archaeology Society.
Published by Blackwell Publishing Ltd. 9600 Garsington Road, Oxford OX4 2DQ, UK and 350 Main Street, Malden, MA 02148, USA.

NAUTICAL ARCHAEOLOGY, 37.1

Figure 1. The Siljan waterway, its lakes and tributaries—a possible area for the use of the boat. The find-spot of the Siljan boat is marked with an arrow. (Jostein Gundersen)

Figure 2. The Siljan boat as it is kept today at the Telemark Museum in Skien, October 2006. (Pål Nymoen)

and the sea in the south, near the fjord town of Larvik. This is the Siljan river (Fig. 1). The boat which was found here is made of a single tree- trunk, and fits the definition of a logboat, being a hollowed-out tree-trunk used as a vessel (McGrail, 1978; Philipsen, 1983). Many years later, it was documented that the logboat is in fact one of the oldest boats found in Norway. C- 14 analysis indicates that the boat dates to the

Roman era, c.240+/−70 AD. Although the Siljan boat is made of a single piece of wood, it is more specialized than a general logboat. Among other things, the boat challenges our notions regarding what a boat is and how it should look (Fig. 2).

Since its discovery, the Siljan boat has not been treated in any way. After the boat was removed, in 1930, from the peat bog that had preserved and protected it, it began drying out, with the

4 © 2007 The Author. Journal Compilation © 2007 The Nautical Archaeology Society

P. NYMOEN: SOURCES FOR NEW NARRATIVES ABOUT RIVER TRAVEL?

Documentation drawing of the Siljan boat. (Knut Paasche and Arne Emil Christensen, 21 January 2005)

Figure 3.

Drawing and photograph showing the preserved ‘handle’ from the Siljan boat. (Pål Nymoen, 8 February 2007)

result that the wood (oak) has now split in several places. In order to secure the continued value of the boat as an archaeological resource, it has now been measured and documented with a 1:1 scale drawing (Fig. 3). Despite the damage caused by the wood drying out and splitting, there are several stretches along the sides of the boat where its contour is completely preserved, making it possible to reconstruct the vessel’s

Figure 5. Ring-ray diagram showing where the Siljan boat was hewn from the original tree trunk. (Pål Nymoen)

original form. Aside from the damage caused by desiccation and the fact that parts of the gunwale are missing, the boat is relatively complete. It seems unlikely that the last owner of the boat abandoned it in the bog because it was damaged.

An interesting detail of the boat’s construction is that there are bow-shaped cut-outs hewn out at each end of the boat just at the transition from the bottom of the boat to the sternpost (Fig. 4). While this part of the boat is broken off at one end, traces of the cutting are still visible, confirming that there were similar cut-outs at both ends of the boat. As the section drawings and Fig. 5 show, the boat was fashioned from one-half of a log split along its length in such a manner that the top of the gunwale and the bow-shaped cut-outs

Figure 4.

© 2007 The Author. Journal Compilation © 2007 The Nautical Archaeology Society 5

NAUTICAL ARCHAEOLOGY, 37.1

Figure 6. The map shows the Hjartdøla waterway, Heddøla, Heddalsvannet and the connection to the town of Skien and the coast. The find-spot of the Hjartdal boats is marked with an arrow. (Jostein Gundersen)

forward and aft were once near the innermost part of the trunk. The boat is both wider and lighter than most of the other logboats from Norway.

It was probably hewn with an iron edge; most probably several different edges or axes were used during the different phases of the forming process. The traces left by the iron tools are still visible, especially on the floor of the boat. The hull is thick at the bow and stern, while the sides are only 1–2 cm thick. The chosen material and manner of construction determined the open form and nearly-flat underside of the boat. While the underside retains the form of the original tree-trunk, it appears that there was an attempt to make the interior appear even and symmetrical. There is no indication that extra planks were attached to the hull, nor that strengthening ribs were used to counteract outward pressure. There is no evidence that an outrigger frame was attached to the boat, which is 5.7 m long and about 90 cm wide.

The majority of logboats found in Norway retain more of the roundness of the original tree- trunk in cross-section than does the Siljan boat. While oak was used to make some of the Norwegian logboats, most of them are made of pine. So far there has been no systematic, comparative study of the Norwegian logboat material. My comments regarding other logboats are based solely upon publications of individual finds (for example Molaug, 1977; Molaug, 1984; Belland, 1986; Fossen,

1992; Christensen, 1996; Arisholm, 1997; Kristiansen, 1997; Carpenter, 2005; Christensen, 2005).

Based upon Per Smed Philipsen’s (1984) com- prehensive study of 224 logboats from Denmark, it appears that the bow-shaped cut-outs at the bow and stern of the Siljan boat are an unusual feature. Philipsen refers to this kind of cut-out as a ‘handle’ and has documented the feature on only two of the Danish logboats. It is interesting to note that a logboat with similar cut-outs has also been found in Ireland (Parret, 1930: 110; McGrail, 1978: 68). I will return to the interpretation of these cut-outs later in connection with my discussion of the use and properties of the Siljan boat seen within the context of the system of waterways in which it was found.

In my search for boats with features that parallel those of the Siljan logboat, I have examined the available publications of logboat finds in Norway. In addition, I have searched the registers of the public institutions responsible for the cultural preservation of vessels. Based upon these investigations, I would like to direct attention to two other boat-finds from the same region as the Siljan boat. As we shall see, these boats have a similar find history and, probably, similar find contexts, properties and uses. The boats differ, however, in age. The two boats were discovered in 1960 in connection with a dam- building project during the construction of the power-plant at Hjartdal in Telemark (Fig. 6). A

6 © 2007 The Author. Journal Compilation © 2007 The Nautical Archaeology Society

P. NYMOEN: SOURCES FOR NEW NARRATIVES ABOUT RIVER TRAVEL?

Figure 7. Documentation drawing of the boat ‘Breidvannet 1’ boat from Hjartdal, Telemark. (Pål Nymoen, 18 October 2006)

Figure 8. Documentation drawing of the boat ‘Breidvannet 2’ boat from Hjartdal, Telemark. (Pål Nymoen, 18 October 2006)

local informant explains that six boats were found and removed when a lake was drained. I have been unable to find out anything more about four of them, but the remaining two are stored on a private farm in the area. A recent C- 14 analysis of the boats, ‘Breidvannet 1’ and ‘Breidvannet 2’, indicates that they date to respectively 990+/−80 years BP and 1015+/−50 years BP (Figs 7 and 8).

So far, there has been no analysis of the type of wood used to build the two boats from Hjartdal, although pine is most likely. The boats were approximately 4.5 and 5 m long respectively, and 55–60 cm wide. As the drawings show, they were also fashioned from half of a tree-trunk; they are thinly hewn and lightweight. However, the bows and sterns are relatively thick. Neither of the boats has any sign of an outrigger, additional planking or ribs. ‘Breidvannet 2’ is quite fragmentary and consists of two main pieces, while the other boat is almost whole. Except for the fact that the two boats were found in the same lake, there are no other details known regarding the circumstances of their discovery. Unfortunately, it is impossible to ascertain whether they were found near land or whether they were intentionally sunk in the lake. ‘Breidvannet 1’, however, was found with a paddle (Fig. 7).

Let us conclude this description of the three logboats from Telemark by raising a question:

what is a boat? Our notions of what constitutes a good boat are tested when it comes to logboats. Conceptions connected to the term ‘boat’ usually include something that floats, something that can transport cargo and people, and the notion that the bigger and more seaworthy it is, the more ‘proper’ it is. Because of these ideas, logboats are often considered to be of low status, both technologically and functionally. However, log- boats are more interesting than such notions would allow, and the Siljan boat, in particular, has something extra. It is both light and elegant with some unusual details. I believe that there are grounds to claim that this boat is an atypical logboat. First, it is thinly hewn and more streamlined in its contours than other known logboat finds. Second, it is older than most of the logboats (with known dates) found in Norway. Last, the bow-shaped cut-outs in the bow and stern are not found on any earlier boat either identified or described in connection with other finds from Norway.

Primitive boats?

As mentioned above, there are a number of premises and expectations regarding the properties and appearance of a boat, underlying our notions of what a boat is. For example, we associate the material object boat with a tool whose properties include the ability to travel over

© 2007 The Author. Journal Compilation © 2007 The Nautical Archaeology Society 7

NAUTICAL ARCHAEOLOGY, 37.1

water. Furthermore, the notion of a vessel’s value is understood in terms of the qualities connected to withstanding and exploiting the wind, currents and waves. It is easy to see that logboats do not score well when such criteria for value are applied. Or do they? The fact is that in Norway the tradition of sea-going keelboats is strong, and inland boats, like flat-bottomed rowing-boats and logboats,areusuallyseenassecond-rank,primitive and peripheral (Klepp, 1984; Christensen, 2000: 164). When logboats, called eiker in some Norwegian historical sources, are compared with other types of boat, such as lapstrake coastal vessels, they appear rather primitive. Such comparisons and attempts at creating typologies are not especially precise, because the respective contexts of the different boats (for example, place of origin and use) are not taken into account.

The history of the field shows that even boat archaeologists, certainly unconsciously, have contributed to the stigmatisation of logboats as primitive (for example Halldin, 1950). In the search for the line of ‘development’ of the clinker technique, logboats are often mentioned as one of two possible origins for clinker-built boats. Thus, logboats are seen as prototypes for a more ‘developed’ type of boat. In this line of thought, logboats with added courses of boards along the sides are seen as more advanced, incorporating improvements to increase their seaworthiness (Christensen, 1966; Crumlin-Pedersen, 1970; Westerdahl, 1989). When the focus of research is on technological origins and change is understood as improvement in a developmental line from simple to complex, logboats appear to be less- developed as boats. The following citation exemplifies this kind of positivistic thinking in which the logboat is presented as the zero-point in the evolution of boats:

When boatbuilders first had learned to attach a row of planks or a ‘course of boards’, it was only a matter of time until they found a way to attach two or more rows of boards. When this stage was reached, the logboat was no longer needed as a starting point. It was steadily reduced in size and remained at the end only in the form of a bottom board or keel, as we know it from later types of boat. (translated from Torgvær, 1984: 47, inspired by Molaug, 1984: 34).

It is mainly this ‘prototype’ perspective, a technological evolutionary perspective developed by Ole Crumlin-Pedersen (1970), which is repeatedly used to explain the invention of the lapstrake (clinker) technique. Based on the evidence of over

30 finds of expanded logboats dating from the 1st century AD to about 1700, Crumlin-Pedersen (2006: 38) (and Fig. 9) compares the introduction of supporting ribs to a ‘magical transformation’. He interprets the introduction of ribs as giving rise to the origin of the idea of the clinker- building technique which is understood as producing a proper boat shape. Crumlin-Pedersen clearly distinguishes between what he calls ‘common’ logboats, that is, those types of boats where the finished boat does not extend beyond the dimensions of the original tree-trunk, and expanded logboats.

Crumlin-Pedersen (2006: 47) has described the Siljan boat as resembling an expanded logboat— possibly an unfinished boat in which the ribs had not yet been added. When it comes to the interpretation of the form, properties and use of the boat, I am rather sceptical as to how much an evolutionary perspective like that of Crumlin- Pedersen contributes. In my opinion, this focus of explaining the appearance of boats and culturally-based boatbuilding traditions from a linear, technological evolutionary perspective has contributed to a reductionist and overly- simplified understanding of the logboat as a boat-type. I do not mean to say that the well- argued technological shift represented by the lapstrake technique (clinker) cannot be traced in the expanded logboats, but the empirical foundation is too thin to make categorical statements here. Some experts have argued that the lapstrake tradition can be traced back to the late-Neolithic period (Østmo, 2005). Recently re-dated remains of a boat from Haugvik in Norway extend the history of the lapstrake technique in the north back to the period between 840 and 420 BC (Sylvester, 2006). The find has parallels to the Hjortspring boat (Sylvester, 2006) from the middle of the 4th century BC (Crumlin-Pedersen and Trakadas, 2003), and it demonstrates that the chronology of the lapstrake tradition in north-western Europe is still rather ‘shaky’.

The fact that ‘ordinary’ logboats have been built and used continuously from the Stone Age to recent times makes an evolutionary interpretation problematic (Fig. 10). Therefore, the method for studying these boats should not be based on interpreting them in connection with ‘the clinker- built boats of the coast’, but rather see them within the context of the terrain and climate as well as the transport and storage possibilities of the region. Seen in light of the landscape in which

8 © 2007 The Author. Journal Compilation © 2007 The Nautical Archaeology Society

P. NYMOEN: SOURCES FOR NEW NARRATIVES ABOUT RIVER TRAVEL?

Figure 9. Ole Crumlin-Pedersen’s ‘magical transformation’ of achieving a proper boat shape by both expanding and extending the tree-trunk beyond its own dimensions. A ‘proper boat’ is here represented with a boat from the Kvalsund find, Kvalsund, Norway, c.700 AD. (after Crumlin-Pedersen, 2006: 37, with permission)

Figure 10. A ‘primitive boat’ in use in recent times: a photograph of a man in a logboat. (after Fossen, 1992, with permission)

© 2007 The Author. Journal Compilation © 2007 The Nautical Archaeology Society 9

NAUTICAL ARCHAEOLOGY, 37.1

the Siljan and Hjartdal boats were found, it appears that these criteria for investigating the form, properties and meaning of these boats may be more productive.

The find contexts within a transport landscape?

The two boats from Hjartdal (Figs 7 and 8) were made and used during the transition from the Viking age to the middle ages. The clinker-building technique was already well established in Norway at this time. There is no reason to think that the people who used these boats were not acquainted with this technique. I base this assertion on the fact that the distance to large lakes with connections to trading centres on the coast was relatively short. Archaeological investigations have shown that whetstones, among other things, were exported by ship in large quantities from these centres (Myrvoll, 1986; Myrvoll, 1991, Nymoen, in press). In England and several places in Scandinavia, whetstones originating from this region of Norway have been found in Viking-age contexts (Moore, 1978; Mitchell et al., 1984). This shows that the extraction and transport of this raw material from the inland waterways to the river delta had an early industrial character and that this export was well established in Viking times—the early- medieval period (Myrvoll, 1986).

The other major natural resource that the region of Telemark is known for is iron. A number of iron-extraction works show that extensive iron production took place here during the period when the Hjartdal boats were in use (Martens, 1971). The areas with the richest iron-ore are located in the marshes, an alpine landscape in the mountains at the source of the waterway system (Dalland, 2001). These areas were relatively inaccessible. At the same time, the demand for this and other inland products in the maritime trading network called for the organisation of a system of transport of pig-iron from the large production-centres in the mountains to the coast (Resi, 1995). We can picture the rivers and lakes as transport arteries from the sources of the iron ore in the highland marshes to the coast. There are, thus, good reasons to suppose that the Hjartdal boats may have had a role in this transport system. Differences in altitude, numerous rapids and waterfalls and relatively-long overland stretches divide the topography of the watercourse into several zones that were navigable by boat. Therefore, it seems likely that these boats were adapted for

use in limited zones within the river system. The types of boats associated with these zones have been described earlier as being particularly suited to the natural conditions and climate, while the zones are defined by the topography and the stretches of water where boats could be used (Westerdahl, 2000: 14ff).

In the county of Aust-Agder in south-western Norway, two presently-undated logboats have been found in a context associated with early iron production (Carpenter, 2005: 81). These boats have not been thoroughly studied. In other words, there is little empirical basis for any assumptions regarding the use of the Hjartdal boats at the present time. Nevertheless, as a hypothesis and a topic for further investigation, it is relevant to consider the boats in connection with the transport of two significant natural resources from the region, whetstones and pig-iron. Because of the obstacles presented by the terrain, it is possible that transport may have taken place in several stages or zones and that different boats were used in the different zones.

Unfortunately, there is little archaeological evidence from the Roman period, the period in which the Siljan boat was in use, that can help us to understand what kind of expectations and demands people had in relation to their boats. Finds of imported goods and materials in grave- mounds suggest that the inhabitants of the area had a close connection to the coast (Mikkelsen, 1976; Vaagland, 2002). The existence of close ties between the inland and coastal areas connected by the same system of rivers is a phenomenon that can be observed throughout Norway. The Skien river system, in part of which the Hjartdal boats may have been used, lies just to the west of Siljan (Fig. 11) and exemplifies this situation. The archaeological landscape along this river system has a noticeable concentration of remains of early settlements along the rivers and around the lakes (Munch, 1962; Kaland, 1969). Finds of imported goods from the same area clearly show that the watercourse provided opportunities for trade and other forms of contact between the inland areas and the coast, and thereby, across the North Sea basin (Myrvoll, 1984; Myrvoll, 1986).

The small lake where the Siljan boat was found lies, as mentioned earlier, in the middle of a system of waterways. Small rivers and streams lead from the small lake, by way of other lakes, to larger lakes in the south. The rivers are connected to one another throughout the area. From the find-spot

10 © 2007 The Author. Journal Compilation © 2007 The Nautical Archaeology Society

P. NYMOEN: SOURCES FOR NEW NARRATIVES ABOUT RIVER TRAVEL?

Figure 11. The two systems of waterways, Siljan and Breidvann, with parts of the Skien watercourse. (Jostein Gundersen)

of the Siljan boat, there is a distance of 24 km to the fjord in the south at the mouth of the low- lying lake, Farris, which lies at a height of 21 m above sea-level today. The land-mass in this region has risen about 8 m in the course of the last 1700 years or so (Henningsmoen, 1979). This means that the difference in the elevation of the outlet of Farris lake and the sea was only about 13 m when the Siljan boat was in use.

This river system extends for several kilometres northward. From the find-spot at Siljan, there are watercourses that, by means of streams, small rivers, lakes and overland stretches, extend far into the mountain regions. The system runs north-south and is separated from two large river systems, Numedals-Lågen in the east and the Skien waters in the west (Fig. 1), by two watersheds. Physical remains of sunken roads have been documented in several places in the district of Siljan (Vaagland, 2002: 21), and it is likely that transit routes also crossed the Siljan river system going east-west even in prehistoric times. At the place where the sunken road meets the Siljan waterway, there is a rather large burial place, probably of the Roman era, as well as a 12th- century stone church. The presence of both these sites indicates that this location was a junction for transit in the area. Evidence for the use of waterways is far less clear than for the use of sunken roads. Wakes left by the boats of the past

have disappeared, and there are few archaeological sources that can be used to understand the meaning that these lines of transit had. While there are harbours, seamarks, rock carvings, cemeteries and place-names that can provide at least an indirect indication regarding the use of the sea, there are fewer such sources relating to the use of inland waterways. In addition, little research has attempted to examine how the earlier inhabitants of inland regions, the locations of their burial-places, and their use of sunken roads, functioned within the context of transit within a system of small rivers.

More general reflections on the possible use of the Siljan boat for travel, transport, fishing and hunting should be seen in the light of the local topography. The definition of what constitutes a good boat is then more dependent upon its intended use, and the uses and topographical conditions themselves set the terms for how the vessel is shaped. If we imagine that a boat should be able to carry its navigator and others over a quiet stretch of river or be used in fishing and hunting in the same waters, then it seems unlikely that a boatbuilder would go to as much effort as must have gone into building the Siljan boat. A more coarsely hewn, and thereby also heavier, logboat, perhaps in pine, would function just as well under such circumstances. Possibly the boat was built to be used for longer trips and under

© 2007 The Author. Journal Compilation © 2007 The Nautical Archaeology Society 11

NAUTICAL ARCHAEOLOGY, 37.1

more varied conditions, as well as for fishing and hunting (Nymoen, 2005).

In addition to attempting to understand the Siljan boat in terms of its possible use and the opportunities and limitations imposed by its topographical setting, it is also useful to consider the boat from a social perspective. If we imagine that a boat should be able to transport both people and cargo over rivers and lakes, and that it should also be light enough to be pulled or carried over land to other waters, then the Siljan boat is not only functional but well adapted to its intended uses within its environment. From the perspective of those who used the boat, we can imagine that it was an important tool. Perhaps there was a degree of status attached to owning such a vessel. The boat gave its owner the ability to transport both himself, other people, a deer- carcass or iron-ore with relatively little effort across long distances. Perhaps the boat was of such importance that the owner became associated with it (Arnold, 1995: 733): an owner who had to be contacted whenever goods were to be transported or people were to be fetched; someone who owned a tool which made it possible to reach the airlock in a beaver lodge, to lay traps and to set up guide fences; or perhaps to carry raw materials from the mountains all the way to the coast—and return with salt?

The three logboats raise many questions. Our modern view of what a boat is limits our ability to answer them. Still, if we ask ourselves how the terrain was at the find-spot and what kinds of opportunities owning the boat gave to the owner as opposed to not owning the boat, then several new questions and interpretations arise. Especially with regard to the Siljan boat, there are a number of features which indicate that it was built to be used over long distances in the system of waterways. In this respect, the most important features of the Siljan boat are the building-material, its weight, and the cut-outs in the bow and stern. By considering the design of the boat and the extent and location of the water system instead of focussing solely upon the boat in terms of a typology and chronology of boat-types, the use of the term ‘primitive’ to describe the boat is inappropriate. It is possible that the boat is unusually sophisticated: perhaps the bow-shaped cut-outs at each end are specifically adapted to the landscape in which the boat was built and used. Might they be handles that were to be used when pulling the boat, and might they be features as important as the boat’s buoyancy?

Multi-functional boats?—an interpretation

The person or persons who made the Siljan boat were probably quite careful to exploit the potential of the material and to fashion the boat to meet the conditions of the landscape and the types of use. Did they intend to build a boat that would work equally well on water or on land ? (The Norwegian phrase, like godt til vanns som til lands, comes from a popular folktale.) Similar questions may be raised to examine the functional properties of boats from other periods. The Bårset boat, a clinker-built boat found on Nord- Kvaløya in the county of Troms (Gjessing, 1941) and dated by dendrochronological analysis to c.800–895 AD (Pedersen, 2002; Wickler, 2002), may serve as an example. The preserved fragments show that it was a light, easily-steered rowing- boat equipped with eight or nine pairs of oars, making it especially well suited for use on the fjords. It is likely that this type of boat was built in such a manner so that it could be carried over the numerous narrow isthmuses along the coast (Nymoen, 1997). The significance of including portages (drageid in Norwegian) in studies of transport zones, both as transit points within such zones and as catalysts for the adaptation of vessels, is well documented and has recently come into greater focus (Westerdahl, 2000; Westerdahl, 2006). Norwegian sagas, among other sources, record a number of examples of the use of portages. Here, the sources usually describe the use of isthmuses (eidene in Norwegian) along the coast or along inland waterways in connection with the desire to avoid a dangerous stretch of water or as a tactic to surprise an enemy in battle. Another description that gives a better picture of how the light logboats may have been used comes from the Norwegian hunter and explorer Helge Ingstad’s stories of hunting expeditions in a canoe in Alaska:

It was important to find the easiest way through the terrain with the heavy equipment. Rivers and waterways were the least demanding and, where the elevation and distance to the next lake or river was shortest, the vessel and the cargo had to be carried, the so-called portages (translated from Ingstad, 1937: 37).

Nikolay Makarov has conducted thorough archaeological landscape analyses in northern Russia. Here he finds that the use of portages has long been a central element in transit along the waterways of the northern areas (Makarov, 1994). In Norway, there are many areas where both the

12 © 2007 The Author. Journal Compilation © 2007 The Nautical Archaeology Society

P. NYMOEN: SOURCES FOR NEW NARRATIVES ABOUT RIVER TRAVEL?

Figure 12. Aerial photograph of the lower part of the Siljan watercourse with parts still under ice. Skagerrak is visible farthest out and to the right. (Dag Jostein Andresen)

place-name and the features of the landscape indicate that the place has been used as a portage, especially along watercourses. According to Rygh (1898), the name eid originated on farms with an overland pass between two lakes or waterfalls and rapids in a river that boats had to be dragged over (Rygh, 1898: 48). The eid-names allude to places that had to be crossed or a landscape with water on both sides. The name Eid is associated with several locations along the two watercourses in which the Siljan and Hjartdal boats may have been used.

Finds of boats are among the archaeological sources that are most relevant for a study of the portages along the inland waterways of Norway. Research examining the relationships between boat-types, landscape and use of isthmuses has resulted in new insights into boat-based travel and transport along inland waters (Nymoen, 1997; Westerdahl, 2006). The Siljan boat has much to offer to this kind of research. As mentioned earlier, evidence indicating how the boat was used is apparent in the form of the boat itself, as for example, the handles at each end.

In order to document the find-spot, establish a basis for further study of the topography and estimate the possible radius of use for a boat like the logboat from Siljan, a series of aerial photographs over the entire river system was taken, in late winter when the ice had started to break up. Looking at these photos, it occurred to

me that a lightweight boat of this type could have been built for use as both sled and boat. The highest parts of the river are still frozen (Fig. 12), and it is likely that the ice is thick enough to bear the weight of one or more persons and a light boat. Further down, about midway in the Farris lake which runs into the sea, a belt of unstable ice is discernible in the photo. This critical point in the trip could have been managed by having a light boat that could be pulled over the ice and thereafter used as a boat on the open water to and from the sea. This hypothesis is based on a visual analysis of the landscape and climate and is only mentioned here as a topic for further study.

To summarize this interpretation, we can say that the Siljan and Hjartdal boats were relatively light boats. They could negotiate both shallow stretches of river and small streams. For the transport of iron or unworked whetstones, for example, it was important that the hulls were shallow and that they were relatively flat on the underside. Boats fashioned from a single tree- trunk, despite the thinness of their hulls, are stronger than boats built with boards. I have had the opportunity to try out a couple of logboat replicas in a river. The logboats were more stable than I had expected and it seemed that the stem and stern took the brunt of the strain since they often hit stones in the shallows. Of course, these paddling experiments are not representative or

© 2007 The Author. Journal Compilation © 2007 The Nautical Archaeology Society 13

NAUTICAL ARCHAEOLOGY, 37.1

scientific, but just the same, the experience provides food for thought. The theory that the boats were used over great distances along the waterway all the way to the outlet to the sea may explain the need for boats that could stand up to difficult and varied conditions, such as shallow and stony stretches of rivers and streams. These conditions may explain the relatively sturdy stems of the Hjartdal and Siljan boats.

This interpretation proposes that the three boats may have been used across great distances, probably in relation to seasonal activities. I think that the connection to seasonal activities is especially relevant for the Hjartdal boats. Because of the topographical conditions along the waterway, the boats may have been made for use in more limited stretches and several boats, specifically adapted for these stretches, may have been needed. The climate and the need for storage in these areas requires boats that could tolerate storage under ice. Boats built with boards are more fragile, so the Norwegian winter, combined with use in shallow rivers, may be a factor in explaining why logboats were used on inland waters from the earliest times up to our own.

Are there more?

In Norway, there are many areas with small watercourses that are topographically comparable to the areas in which the Siljan and Hjartdal boats were found (Fig. 11). If we examine the maps with the idea that rapids and isthmuses do not represent barriers to the use of small, light boats, there are many places where we might expect to find such boats. Why, then, have we not found more of them?

The first, most obvious, explanation is that there have been few attempts in Norway systematically to register finds of cultural material under water in inland waters. Further, a quick look at the finds of logboats registered at the Norwegian Maritime Museum reveals that all of them were found by chance. Finds made in connection with flooding, the draining of lakes and waterways, digging associated with building or public works projects, the salvage of sunken logs, and fishing excursions, are examples of what has been recorded under the heading ‘find circumstances’ in this register.

The other explanation may depend partly on how logboats were kept. The most secure way to prevent a logboat from splitting and from other types of damage is probably to ‘store’ the boat

by filling it with stones and sinking it in water (Christensen, 1996: 240). What then happens when the owner does not come back to recover the boat one spring? What happens after a few years when the bog becomes overgrown? When no-one looks for them, they are found only by chance, as in the case of the Siljan boat. Once, when I was surveying the find-spot, I was told that the find was made by chance when a chunk of peat was turned up, revealing the boat. It is often coincidental both that a boat is found and that it is then kept. Perhaps it is not so strange that so few logboats have been recorded?

Our view of prehistory, interpreted through the archaeological record, is only so wide as the sum of the remains we have recovered. Archaeologists have not searched for logboats in the areas in which they were usually used. The reason for this is that many of them lie under water, and underwater archaeology is seen as a relatively new method in terms of the history of the field— even in Norway.

Sources of new narratives?

The archaeologist’s view is limited, and an increased effort to register archaeological material associated with inland waterways can make an important contribution to our understanding of this little- studied part of prehistory. The close ties between people and water—and the use of water—are both obvious and acknowledged in archaeology, particularly in relation to themes like settlement patterns, resource exploitation, trade, transit, rock art and other symbolic uses. When archaeological narratives about these themes are written based mainly upon sources that have been deposited on land, or when the narratives we make about transport and transit on water focus exclusively on the maritime sphere, the narratives are incomplete. How would the picture we have of prehistory look if the survey and registration of cultural material from lakes and waterways was as thorough and comprehensive as it is for material found on land? ‘Maritime archaeologists’ must therefore shift their focus inland to a greater degree—where there is clearly great potential to find boats from periods that are less well-known and to shed light on themes that are little understood. In my opinion, such a broadening of scope would raise new questions and open up new evidence and new perspectives for research on how people used water as well as on their relationship to water.

14 © 2007 The Author. Journal Compilation © 2007 The Nautical Archaeology Society

P. NYMOEN: SOURCES FOR NEW NARRATIVES ABOUT RIVER TRAVEL?

With the exception of research into the origins of boat-types and evolutionary perspectives on boat development, scholars have mainly left logboats in peace. Since we know something about how long logboats have actually been in use and how they have been used over large areas with varied terrain, it seems that the evolutionary perspective is not suited to create narratives about these boats (Nymoen and Arisholm, 2005). My view is that we will find new narratives about the oldest boats by looking at the systems of waterways that connect the inland regions with the coast. Some of the new narratives may be about travel, extent, and organising of transport between these areas. They might also enlarge the picture of ‘sea routes’ linking two coasts to include extensions of these routes far into inland regions. Perhaps these stories will also tell about boats that function both in coastal and inland regions?

Acknowledgements

It seems that logboats are a type of archaeological source to which we only coincidentally have got access. The find circumstances clearly attest to this fact. With the exception of grave finds, there has not been a single logboat found as part of a systematic survey or any other archaeological investigation in Norway. Because logboats are best kept when they are under water, and because many of them remained under water when they ceased to be used, it is likely that even very old logboats are still preserved—and hidden. From this we must conclude that sources of knowledge about the use of lakes and inland waterways in prehistory exist but that they are little used in research. Close study reveals that logboats are far from a homogeneous type of vessel and that they do not follow a line of technological evolution. Rather, logboats represent a multitude of different vessels, not only in terms of time and space, but also in terms of their functions.

I would like to thank Lisa Benson for translating the text from Norwegian to English. In addition, I would like to thank Lars Vaagland of the Siljan Historical Society and Halvor Åbø, Sverre Haugan and Ambros Langåsdalen in Hjartdal for assistance and information about the logboats. Thanks are also due to Dag Jostein Andresen who took the aerial photographs, Linda Marie Rustad for commenting upon the text, and Jostein Gundersen for the GIS maps.

References

Arisholm, T., 1997, Norges eldste båt opp av mudderet etter 2000 år, in H. Sørheim (ed.), Arkeologi og kystkultur, 105–13. Ålesund. Arnold, J. E., 1995, Transportation Innovation and Social Complexity among Maritime Hunter-Gatherer Societies, American

Anthropologist 97.4, 733–47.
Belland, S., 1986, 1000 års gamal stokkebåt funnen i Haukomvatnet, in Birkenes historielags årsskrift. Birkenes.
Carpenter, S., 2005, Fra stokkebåt til flatbunning?—En vurdering av materialet fra Aust-Agder, in T. Arisholm and P.

Nymoen (eds), Stokkebåter—Nytt om Sørumbåten og andre sørnorske stokkebåtfunn. Norsk Sjøfartsmuseum. Skrift nr 4,

69–83. Oslo.
Christensen, A. E., 1966, Frå vikingskip til motorsnekke. Norsk Kulturarv 2. Oslo.
Christensen, A. E., 1982, Viking Age Ships and Shipbuilding, Norwegian Archaeological Rewiew 15.1–2, 19–28.
Christensen, A. E., 1985, Boat finds from Bryggen, The Bryggen Papers, Main series Vol I, 47–278. Bergen.
Christensen, A. E., 1996, Innlandsbåtene, spennende men lite kjent, in T. Fossum (ed.), Årbok for Norsk Skogbruksmuseum

14, 229–46.
Christensen, A. E., 2000, Some archaic details of Norwegian fresh-water boats, in J. Litwin (ed.), Down the River to the Sea.

Proceedings of the 8th ISBSA, 163–8. Gdansk.
Christensen, A. E., 2005, Stokkebåt, Norsk Arkeologisk Leksikon 362. Oslo.
Crumlin-Pedersen, O., 1970, Skind eller træ. En studie i den Nordiske plankebåds konstruktive opprinnelse, in Hasslöf et al.

(eds), Sømand, Fisker, Skib og værft. Introduksjon til maritim etnologi, 213–39. København.
Crumlin-Pedersen, O., 2006, Den nordiske klinkbåds grundform—en totusindårig tradisjon og dens rødder, in T. Arisholm,

K. Paasche, and T. L. Wahl (eds), Klink og seil—Festskrift til Arne Emil Christensen, 36–56. Oslo.
Crumlin-Pedersen, O. and Trakadas, A. (eds), 2003, Hjortspring. A Pre-Roman Iron Age Warship in Context. Ships and Boats

of the North 5, Roskilde.
Crumlin-Pedersen, O., 2006, Den nordiske klinkbåds grundform—en totusindårig tradisjon og dens rødder, in T. Arisholm,

K. Paasche, and T. L. Wahl (eds), Klink og seil—Festskrift til Arne Emil Christensen, 36–56. Oslo.
Dalland, Ø., 2001, Telemark—i lys av vannet. Oslo.
Fossen, A., 1992, Historien om de Norske skogfinnene. Oslo.
Gjessing, G., 1941, Båtfunnene fra Bårset og Øksnes. Tromsø museums årshefter 58, Tromsø.
Halldin, G., 1950, Bilder av primitiva farkoster m.m. sammanställda för studiet av de skandinaviska hällristningarnas

skeppsbilder, Sjöhistorisk Årsbok, 11–96. Stockholm.
Ingstad, H., 1937, Pelsjegerliv. Blant Nord Kanadas Indianere. Oslo.

© 2007 The Author. Journal Compilation © 2007 The Nautical Archaeology Society 15

NAUTICAL ARCHAEOLOGY, 37.1

Henningsmoen, K. E., 1979, En karbon-datert strandforsyvningskurve fra søndre Vestfold, in R. Nydal, S. Westin, H. Ulf, and S. Gulliksen (eds), Fortiden i søkelyset. 14C datering gjennom 25 år. Trondheim.

Hougen, B., 1941, Islendingen i eika, Viking V, 51–74.
Kaland, S. H. H., 1969, Studier i Øvre Telemarks Vikingtid, Universitetets Oldsakssamling Årbok, 67–216. Oslo.
Klepp, A., 1984, den styggeste av alle vore farkoster ... en undersøkelse av flatbunnede båter i Norge, Norsk Sjøfartsmuseum

Årsberetning 1983, 67–127. Oslo.
Kaland, S. H. H., 1969, Studier i Øvre Telemarks Vikingtid, Universitetets Oldsakssamling Årbok, 67–216. Oslo. Kristiansen, F., 1997, Nytt stokkebåtfunn innrapportert til Vitenskapsmuseet, Trondheim, Spor 1, 23. Trondheim.
Kvalø, F., 2004, Facing the sea in Bronze Age Norway: the ship, the sea and society, in P. Clark (ed.), The Dover Bronze Age

boat in context: society and water transport in prehistoric Europe, 148–52. Oxford.
McGrail, S., 1978, Logboats of England and Wales. BAR Brit. Ser. 51; I and II. Oxford.
Makarov, N., 1994, Portages of the Russian north: Historical geography and archaeology, Fennoscandia Archaeologica XI,

13–27. Helsinki.
Martens, I., 1971, Møsstrond i Telemark. En jernproduserende fjellbygd før svartedauen, Viking 36, 83–114.
Makarov, N., 1994, Portages of the Russian north: Historical geography and archaeology, Fennoscandia Archaeologica XI,

13–27. Helsinki.
Mikkelsen, E., 1976, Arkeologiske undersøkelser i Siljan-vassdraget. Oslo.
Mitchell, J. G., Askvik, H., and Resi, H. G., 1984, Potassium-argon ages of schist honestones from the Viking Age sites at

Kaupang (Norway), Aggersborg (Denmark), Hedeby (West Germany) and Wolin (Poland), and their archaeological impli-

cations, Journal of Archaeological Science 11, 171–6.
Molaug, S., 1977, Stokkebåt fra Kroktjern Søndre Fange, Norsk Sjøfartsmuseums Årsberetning 1976, 13–14. Oslo.
Molaug, S., 1984, Utriggerkano og pram fra Holmsbu, Norsk Sjøfartsmuseum årsberetning 1983, 129–5. Oslo.
Moore, D. T., 1978, The Petrography and Archaeology of English Honestones, Journal of Archaeological Science 5.1, 61–73. Myrvoll, S., 1984, Trade in Telemark and the Earliest Settlement in Skien, Offa Band 41, 41–55. Neumunster.
Myrvoll, S., 1986, Skien og Telemark—naturressurser, produkter og kontakter i sen vikingtid og tidlig middelalder, Viking

1985/86, 161–80.
Myrvoll, S., 1991, Skien i Grenland—om kommunikasjon, økonomi og makt i sen vikingtid og tidlig middelalder, Gunneria

64.2, 289–301. Trondheim.
Munch, J. S., 1962, Borg og bygd. Studier i Telemarks eldre jernalder, Universitetets Oldsakssamling Årbok 1989/1990, 7–175.

Oslo.
Nielsen, Y., 1905, Middelalderske samfærdselslinjer i Norge, langs kysten og paa indsøer og elve, Norske geografiske Selskabs

Aarbog XVI. Kristiania.
Nymoen, P., 1997, Der er en eiendommelighed ved den Norske kyst—Om eidenes betydning for ferdselen til sjøs, Kysten 1,

16–20.
Nymoen, P., 2005, Like godt til vanns som til lands ?—En stokkebåt fra Siljan i Telemark, in T. Arisholm and P. Nymoen

(eds), Stokkebåter—Nytt om Sørumbåten og andre sørnorske stokkebåtfunn. Norsk Sjøfartsmuseum. Skrift nr 49, 57–68.

Oslo.
Nymoen, P., in press, Fatal feilnavigering med tung last ?—om en samling brynesteinsemner ved Kvaasefjorden, Kristiansand. Nymoen, P. and Arisholm, T., 2005, Oppsummering, in T. Arisholm and P. Nymoen (eds), Stokkebåter—Nytt om Sørumbåten

og andre sørnorske stokkebåtfunn, Norsk Sjøfartsmuseum, Skrift nr 49, 125–7. Oslo.
Østmo, E., 2005, Over Skagerak i steinalderen. Noen refleksjoner om oppfinnelsen av havgående fartøyer i Norden, Viking

LXVIII, 55–82.
Parret, O., 1930, Die Einbäume in Federsee-ried und im übrigen Europa, Prähistorishe Zeitschrift 21, 112.
Pedersen, W. M., 2002, Bårsetbåten. En revurdering av rekonstruksjonen fra 1937. Upublisert hovedfagsavhandling i arke-

ologi, Institutt for samfunnsvitenskap, Universitetet i Tromsø.
Philipsen, P. S., 1983, En analyse av Danske stammebåde. Upublisert hovedfagsoppgave i arkeologi, Århus universitet.
Resi, G. H., 1995, The Norwegian iron bar deposits: have they most to tell about production or consumption, Varia 30, 131–

46. Oslo.
Rygh, O., 1898, Norske gaardsnavne, Forord og Indledning. Kristiania.
Sylvester, M., 2006, Haugvikbåten fra Sømna—en plankebygd båt fra yngre bronsealder eller førromersk jernalder, Viking

LXIX, 91–106.
Torgvær, T., 1984, Fra trestamme til nordlandsbåt, Leddiken Årg. 5. Øksnes.
Vaagland, L., 2002, Den hemmelighetsfulle dalen, Kult makt og ære i Siljan—Fra Skrim til Farris, bind VI, 7–73. Siljan

Historielag. Larvik.
Westerdahl, C., 1989: Norrlandsleden I. Kallor till det maritima kulturlandskapet. Länsmuseet–Murberget.
Westerdahl, C., 2000, From Land to Sea, from Sea to Land. On the transport zones, borders, and human space, in J. Litwin

(ed.), Down the River to the Sea. Proceedings of the 8th ISBSA, 11–20. Gdansk.
Westerdahl, C., 2006, On the Significance of Portages. A survey of a new research theme, in C. Westerdahl (ed.), 2006: The

Significance of Portages. BAR Int. Ser. 1499, 15–51. Oxford.
Wickler, S., 2002, Båtkonstruksjon i Nord Norge, accessible at: http://www.imv.uit.no/ommuseet/enheter/ark/

forskning_Botkonstruksjon.htm
Østmo, E., 2005, Over Skagerak i steinalderen. Noen refleksjoner om oppfinnelsen av havgående fartøyer i Norden, Viking

LXVIII, 55–82.

16

© 2007 The Author. Journal Compilation © 2007 The Nautical Archaeology Society