The Retinue and the ShIP An Archaeo-sociological Study of Scandinavia at the Tum of the E , ast Millennium and the Following Centuries

administrative space introduced and explained to those who belonged to a realm of power? We have to remember that such a project had to be made comprehensible to a society that did not have the spatial and administrative frame of reference that we do. Here too, the principle of similarity, or rather recognition, is a key to understanding. In order to change, continuity was essential. This is obviously a contradiction, but let us not forget the zeitgeist. Ideally, nothing neither could, nor should change, but as we all know, it always does. Using Gidden's ideological terminology, the intentions and interests of the sectional group had to be presented as universal, and contradictions had to be hidden (or masked) (1979:193ff).As Gilles Deleuze put it: "a new force can only appear and appropriate an object by first of all putting on the mask of the forces which are already in possession of the object" (1983:5). As mentioned above, in order to take central control the medieval Scandinavian kings had to solve a three-fold problem, and the time has come to examine this issue a little closer: How to transmit incomes from landowners to the crown. How to monopolize the use of force. How to explain and legitimize central administration, in people's mind and in


ARCHAEOLOGICAL LANDSCAPES
Examining landscapes in more or less pragmatic terms has been an important part of archaeology for several decades, but interest in how the idea of landscape is socially constructed has increased considerably during the last five years (e.g. Barrett 1994;Tilley 1994;Hirsch k 0'Hanlon 1995;Bradley 1994Bradley , 1997Bradley , 1998Johansen 1997;Nash 1997). In this paper, I will discuss a model for the interpretation of the social landscape of Viking Age/early medieval Scandinavia which is based on a more extensive work (Varenius 1998). For obvious reasons I have been forced to limit myself to certain aspects and to omit others, for example omitting lengthy quotations of the provincial laws, as well as the rather internal Swedish discussion concerning the formal character of the settlement units during the period. Here, analysis focuses on the use of metaphorical analogies, but I hope to be able to demonstrate that the metaphorical thinking is not only an analytical tool, but also something which carries an actual historical authenticity.

A SOCIAL THEORY
Historians and archaeologists have long since agreed that the Scandinavian society around and after 1000 AD was clearly stratified, both economically (=resources) and socially (=position), and today this opinion seems selfevident and in no need of further argument (c.f. Norr 1998;Zachrisson 1998).The usual way of acknowledging this has been to categorize the population into a restricted set Currem S«:e&/is/~Archaeotogi:, Voi. 7. /999 of socially defined groups, for example kings and royals, wealthy landowners, so-called normal peasants, eventually with further subgroupings. These social strata, similar to political classes, have also been sought for, and hence encountered, in the archaeological material (e.g. Broberg 1993:245f). But as Heikko Steuer (1982) has argued, the social inequality in the Merovingian and Viking periods may not be so much a question of defined and clearly separated classes. Instead, he talks about a Rang-Gesellschaft that was composed of so-called familia. groups with a common structure and shared societal ideals, but with different material and social prerequisites. Of course there were great differences among familia.that is not an issue of debatebut along a common line of organization. The wealthier had moreand more expensiveitems and property, and were socially (and perhaps genealogically) closer to the royal power. Yet, there seems to have been a structural similarity, a web which held together large parts of the social hierarchy, and such features would be a mistake to forget. They may help explain some of the characteristics ofthe archaeological material.
Parts of this basically Continental social model also have relevance for Scandinavia.
It seems to me that the structure of the family the householdalso functioned as a typification of society.  Sawyer (1988) has pointed out that women are mentioned in about 30% of the stones, which means that in 70% of the stones only men occur. Thus the stones as a medium are male dominated, and this is even more obvious when considering that the frequency of stones without men amounts to less than 1%.
Thus a basic social scheme based on male relations can be identified; its three most common relationships are fatherson, brother brother and sonfather ( fig. 1). There can be no doubt that this was an ideal model, which means that in people's social reality these roles were not always played by men.
Sometimes they were also performed by women, doing what men were doing and/or doing things that men did not do, but in spite of this, the structure was essentially designed by men, for men. Perhaps this can be described as the roots of androcentrism. The structure can also be traced in scaldic poems, sagas etc. , testifying to nearly patrilineal visions of how society should be organized, although nowadays it is generally accepted that Scandinavian kinship was bilaterally recognized, as in the rest of western Europe (Murray 1983).Inherent here was also an idea of power, a sense of domination, where a man was subordinate to his father but in charge of his sons, hence a vertical line of domination.
Regarding a man's relations to his brothers, to whom the links were horizontal, there were  2).
However, relationships between male family members, i.e. beween fathers, brothers and sonsno matter their importance in their own rightseem to have had an even greater significance on a metaphorical level. The same sources use the family member terminology in other ways too, namely when referring to people to whom the subject was not genetically, but emotionally and socially, related. Perhaps this is most obvious in the retinueinstitution, the licl, where comrades are described as brothers(-in-arms), as fatherlike, or son-like. To judge from the various comments and references we can find among rune stones and the Norse literature, male family relationships were used as a means of creating social roles. The tools by which others were identified were domestically defined; they were recognized in the same terms as the members of the family. An obvious advantage of this was that the praxis of behaviour was not a big problem, once the relationship was settled ( fig. 3). From the works ofPierre Bourdieu, we can see that this was not a unique phenomenon. The Catholic church is another sphere where this symbolic "family spirit" flourishes, and here too it leads to hidden but firm forms of domination (Bourdieu 1998:102-122).This mode of expected behaviour may be described as a case of habitus.
We can deduce from the various sources that the retinue was hierarchically organized, meaningat least in theorythat the person at the top could control an ever increasing number of people, depending on how many followers he had. In tum, these followers probably also had followers, just as overlords controlled a number oflords (Sawyer 1991:8). This is why alliances were of such vital importance for kings and other leaders. One of the most renowned examples of this eAect is the story of Ingvar and his expedition to "Särkland", known both from (the rather  5). Both sources bear witness to that the expedition was a disaster, and that both Ingvar and a great deal perhaps mostof his men died. Regardless of this, we can discern two interesting phenomena here. First we can learn the principle of hov such an expedition may have been recruited. Seen from the assumed center, stones occur in all directions, most vastly spread to the north and to the south, in both cases some 100-120 km. Unlike some other scholars (c.f. Larsson 1986;Gahrn 1989), I do not see an underlying administrative organization as responsible for the recruitment ofthose men. To me, the pattern is much more likely to be a result of rumour, word of Ingvar and his intentions being spread in more densely populated areas. I think the main reason for those men to join Ingvar was their own ambition, because such expeditions were part of a social ideal. You would want to participate. The second thing we can learn from this example is that failure seems to have had no negative effect on the memory of Ingvar and his men. On the contrary, such a splendid disaster may have been the single most important factor that helped preserve the legend of Ingvar. Success would probably have given the participants prestige and some material wealth in their own lifetime, but their shortcomings made them immortal and forever remembered (fig. 6).
The stones demonstrate what I believe was an ideal social order: family relations being a model for how to behave towards each other; entire units could be added without problems because of the established hierarchical principle of dominance; and men of different background and material wealth shared and fulfilled the same societal values to the best of their capabilities. CONSTRUCTING A TERRITORY Today, almost as a truism, we regard Scandinavia in the 11",the 12",and to some extent also the 13'" century, as the scene of both ongoing conftict and rapid change. Though this certainly is a cliché in many respectswhat part of history is peaceful and static?it is obvious that the early medieval kings and other aspirants to formal power fought for supremacy in a manner that has left quite a number of traces in the historical records. Eventually it was a matter of establishing central control, or at least central influence, over local and regional bureaucracy concerning legislation, commerce, religion etc.
Three goals appear as especially vital for the earthly potentate-to-be: the monopoly on the use of force, the creation of a system for transmitting incomes from landowners to the crown (i.e. taxation), and finally, the establishment and legitimization of central power.
Early medieval kings did not rule formal countries, they did not rule territories with fixed and acknowledged borders as are the criteria today; that is also something that we "know" by now (c.f. Wallerström 1995:20ffl. "Countries" or "states" are thus unsuitable terms; rather, we should think in terms of "people" in a certain region (riki), but as an effect of this, the region is associated with that group and also connected to it by name. There are several runic inscriptions in which this relationship is manifested, the most notable being the big Jelling stone: sa Haraldr ces seR wan Danmorl' . . . ok dani gcerPi Kristna (that Harald, who won the whole of Denmark . . . and made the Danes Christians). A Swedish stone (Sm 1) with a similar meaning says: Bonda virclskum hiogg As- property, or as Dagfinn Skre expresses it: "the lordship over things was linked to the lordship over human beings" (my translation) (Skre 1997:17).This sounds perhaps quite feudal, which it may well be, but the point is that the question of owning was not restricted to an economic transaction. It was also a matter of control, to be a part of, to be in possession of. Also people wereor could bepossessed or "possessed over", and wanted to be, as part ofbeing-in-the world. In a parallel but related line of thought, concerning the mentality of odal property, Aron Gurevich has stated: "just as the man possessed the farmstead . .. it in tum 'possessed' him and put its own stamp on his personality" (1985:48).
My conclusion of this brief discussion of people and land is that "belonging" was important; you belonged to a farm and you belonged to a family household. You also belonged to someone. As I see it, being "free" was not an advantage, if you by that mean free in the modern sense: that you can do what you want, no one can interfere with your business, no one cares about you, and no one expects you to do anything for them. On the contrary. For the Viking Age family, it was most important to belong to a powerful man's or woman'srealm of interest, but it would be wrong to see that as a forced subjugation. Social relations in both horizontal and vertical direction were rather the bonds that guaranteed everybody's safety and that everybody had their place in societyalthough some for better and some for worse. Being outside the community was to be without support, and being without support was to be without peace. Social legitimacy was to be within.
What must be explained and understood then, is how this socially constructed landscape became territorialized. Not that people were without notions of borders, regions and other kinds ofcultural and/or legal boundaries before, but what I am referring to is the idea ofadministrative territories. How was the idea of formal duty presented, by what arguments were people convinced to pay tribute and to do service to an abstract organization? How did the principle of regionality become equally important, and eventually more Current Swedish Archaeotogy, Vol. 7, l999 important, than the principle of personal belonging? How was the idea of belonging to abstract, administrative space introduced and explained to those who belonged to a realm of power? We have to remember that such a project had to be made comprehensible to a society that did not have the spatial and administrative frame of reference that we do.
Here too, the principle of similarity, or rather recognition, is a key to understanding. In order to change, continuity was essential. This is obviously a contradiction, but let us not forget the zeitgeist. Ideally, nothing neither could, nor should change, but as we all know, it always does. Using Gidden's ideological terminology, the intentions and interests of the sectional group had to be presented as universal, and contradictions had to be hidden (or masked) (1979:193ff).As Gilles Deleuze put it: "a new force can only appear and appropriate an object by first of all putting on the mask of the forces which are already in possession of the object" (1983:5).
As mentioned above, in order to take central control the medieval Scandinavian kings had to solve a three-fold problem, and the time has come to examine this issue a little closer: How to transmit incomes from landowners to the crown.
How to monopolize the use of force.
How to explain and legitimize central administration, in people's mind and in space.
Naturally, this was not done with a single technique all over Scandinavia, and not at the same time. What interests us here is a special case that involvs strong use of maritime metaphors, and that for obvious reasons was most applicable in the coastal regions. In addition, I believe it can best be understood as an attempted strategy, one that was only partially successful.
In several provincial laws in all three Scandinavian countries there occurs the ledung organization, not identically described but with considerable structural similarities. The ledungen was a legally grounded institution that provided the idea of a royal seaborne retinue, with the right and duty of all landowning men to serve the king at his will (Lund 1996;Varenius 1998). These laws describe in varying detail how the ships were built, maintained, manned and equipped. Common to them all is that, despite the long timespan of the codiflcation of the laws, ca 1150-1325 (Norseng 1987), the ledung obligations have always just been altered, from the duty to participate in royal expeditions at sea to the duty to pay taxes. What the laws try to explain to the world is the scenario that there for some reason will be no expedition, but that foodstuffs should be delivered to the king's warehouse anyway and money paid in compensation for not joining the cancelled trip. It seems almost unavoidable to draw the conclusion that the main purpose ofthe whole enterprise was to make an excuse for the king to claim various services. This was a taxation system, explained with a historical reference to plunder and/or naval defence.
Calling out the "fleet" was a regal prerogative, hence the monopoly on the use of force. In Norwegian and Swedish laws, the king is more visible in the texts, but also in Denmark the king decided over how to dispose of the resources. Whether or not such a fleet of light, Viking-like ships actually had a strategic importance in the 13'" century may be disputed, but apart from that and other questionable features, it was made clear that the king was the sole commander. I interpret that as a statement that the former habit of keeping private, armed retinues as the basis for personal enterprising was no longer permitted. By swearing an oath to the king, the great landowners and other chiefs had put themselves and their men under his command, but at the same time the king had become dependent of their cooperation and local influence. The means of forceships, Current Sw:edish Archaeology, Voh 7, l 999 8)örn Varenins Fig. 7. The organisation of' the ledungen ship s team (Sw. skeppslag) according to several Swedish provincial laws. Each province should provide the king with a ce&tain numbe& of such ships. By the time the laws were wri tten (ca 1300), these "ships" served (or were meant to serve) the purpose of' taration, th»s c&eating both a social unit and a territorial unit. manpower, horses etc.still very much remained in the hands of the local aristocracy.
The third role of the ledung system was to promote the idea of a ship and a ship's team as a metaphor for the local societythe community. Individual farms and villages were clustered into groups called "oarlock" (Sw. hamna), obliged to put one rower (Sw. håså te) at the king's disposal. A group of such oarlocks made up a quarter, two quarters a half, and two halves a full "ship's team" (Sw. skeppslag). Thus, a ship's team was both a social unit and an administrative unit, hierarchically organized, situated in an actual landscape, and steered (i.e. governed) by a "helmsman" (Sw. styrman), who was also responsible for the collection and delivery of the duties, the taxation ( fig. 7).
In the historical record, bits and pieces of this organization occur, but not in the regular fashion the laws prescribe. Most likely it was never consistently fulfilled, and in those places where it actually was introduced, it was adjusted to local traditions of organization, measuring units etc. Part of that is visible already in the laws themselves. Finally, as to the so-called warfleet question, we have little or no conclusive evidence whether it ever existed, and if it did, what it looked like.
Remains of old warships as reconstructed by maritime archaeologists rarely match the law's standardized naval structure (c.f. Crumlin-Pedersen 1997:92).We can safely assume that the ledungen never operated in the way the laws picture it, and yet it was most useful. It was the means by which medieval Scandinavian kings could make comprehensible their demands without really having to say what they were doing, namely exploiting their own people instead of the enemy.
Curren& Swedish Archaeotogv, Vol. 7, 1999 SUMMARY AND CONCLUSIONS One of the ideas behind this paper was to discuss the social landscape in Viking Agel early medieval Scandinavia, (although most examples here are Swedish). It is certainly not I who has invented that particular metaphor "social landscape", but I like it. However, the idea of "landscape" indicates that social relations are somehow laid out, observable, thus connecting people in various geographical locations in a rather concrete manner. In one sense that may be true, and yet things are not altogether clear: for example, who were connected to each other, how was the raison d'etre of that connection contextually explained (and understood), what changes in the character of the connection can be seen over time? And so forth.
Early administrative terminology "hundreds" (Sw. hundare), "ship's teams" (Sw. skeppslag), "twelfths" (Sw. tolftet)refer both to a social group and a territorial unit. From this, it seems rather natural to deduce that there were close bonds between social structure and administrative organization. A vital component in the historical understanding of this bond is the Viking Age retinue (Sw. lit)). It was organized more or less as a family, with the leader as a "father" and the followers as "sons". Among the warriors themselves there was a strong sense of "brotherhood", especially so in Denmark according to runic inscriptions. An analysis of the well-known Ingvar rune stones in eastern middle Sweden gives us an idea of how a retinue may have been recruited and organized. Several of his sub-chiefs were also ship's captains (styrimadr), they themselves commanding a "lid", most likely what is referred to as a skiblicl.
I think this background is important to bear in mind when turning to the maritime metaphor in the early provincial laws. For once, it is not a matter of plain retrospective projection of a later phenomenon upon prehistory. I wish to allow prehistory to play a part in the formation of the later history, to let society and the mentality of early medieval Scandinavia be affected by what was the historical background and memories of that time. In that way, the pan-Scandinavian levy (ledungen) taxation system, which was presented and explained as a royal war fleet, appears as a conscious attempt to make use of a collective memory of a powerful prehistory.