Stone Enclosures — Linking Time and Guiding Space

From the late Roman Iron Age, stone enclosures and houses with a stone foundation have been built on Gotland. Stone enclosures have generally been interpreted as fences between the infields and the outlying land. I will argue that this explanation is insufficient and that we also have to consider the enclosures' symbolic and social significance. The stone enclosures were part of people's everyday practice as they moved through the landscape, and the enclosures structured the reality. The stone enclosures can be seen as both linking and separating, and one interpretation put forward is that the enclosures "embody" the social structure.

From the late Roman Iron Age, stone enclosures and houses with a stone foundation have been built on Gotland. Stone enclosures have generally been interpreted as fences between the infields and the outlying land. I will argue that this explanation is insufficient and that we also have to consider the enclosures' symbolic and social significance.
The stone enclosures were part of people's everyday practice as they moved through the landscape, and the enclosures structured the reality. The stone enclosures can be seen as both linking and separating, and one interpretation put forward is that the enclosures "embody" the social structure.
The presence of stone enclosures (Sw. vastar or stenstrctngar) all over Gotland has given rise to several interpretations. The interpretations which were put forward in the beginning of the century are often considered imaginative, since they regard the stone enclosures as cultand grave paths or as structures of defence. In the 1960s and 1970s a more functional approach developed, by which the stone enclosures were explained as prehistoric fences to keep the cattle away from the fields.
Stone enclosures are found on the Baltic islands (Gotland and Öland) and on some parts of the Swedish mainland. Especially those in Östergötland have been the subject of a great deal of debate (e.g. Lindquist 1968;Widgren 1983). There are, however, differences between the mainland and the Baltic islands during the early Iron Age, for instance concerning house construction. On the mainland the walls were built of wattle and daub, which makes them hard to find today. On Gotland (and Öland) on the other hand, the houses walls were built of stone, which makes them visible above ground.
Owing to this we can locate the settlements more easily and determine their relation to the stone enclosures.
The point of departure for the following discussion is that the functional explanation, in which the stone enclosures are seen as fences for the animals, is insufficient. Not in the sense that this explanation is "wrong", but because it does not take into consideration that the stone enclosures must have affected people's conception of the world.
The stone enclosures had a symbolic significance and there was also a motive behind the fact that people felt it necessary to emphasize their presence in the landscape with enclosures. This makes the tradition and history of the society an important subject of study. To get a deeper understanding of the stone enclosures we need a more cornplex picture of the society than the one we get by considering only the agricultural technique. In the following discussion I will argue that the stone enclosures are part of people's everyday practice and that they Cnrrent Srvedish Archaeology, Vol. 4, i996 "embody" the social structure of the society.

EARLIER WRITINGS ABOUT STONE ENCLOSURES
It is important to examine previous research on a subject because the interpretations that are today considered as obvious could, for instance, have been produced in a different theory of science. By making visible the different interpretations of a subject such as stone enclosures, it is possible to call into question the prevailing ideas. Such questioning can lead to new ideas and different approaches to the subject. One can say that in the earlier archaeological literature there is a greater variation in the interpretations of stone enclosures than there is today. When Arthur Nordén discussed the stone enclosures in Östergötland, he saw them as a kind of cult path which led towards the graves. He tells us, for instance, about a verbal tradition that explains the stone enclosures as "paths of sacrifice, which they used before, when they sacrificed to their gods and their dead" (Nordén 1930:85, my translation). A similar interpretation was made by John Nihlén, though he did not see the enclosures as a homogeneous group. Nihlén divided them into three groups: i) Graveand cult paths, ii) Settlement-boundaries, farmor village defences, and iii) Fences for the cattle. He thus saw in the stone enclosures both a functional reason and a symbolic significance (Nihlén & Boethius 1933).
They distinguish between enclosures that delimit a four-sided area and which consist mainly of earth, and those that are built mainly of stone and stretch far into in the landscape. They saw the stone enclosures foremost as a result of cultivation and cattle breeding. One should observe that when Dan Carlsson talks about stone enclosures he does not include the earth banks surrounding the fields, which he sees as a result of cultivation in earlier times. Carlsson maintains that the stone enclosures should be seen as fences: "Today there is hardly any doubt that they should be seen as fences or foundations for fences" (Carlsson 1979:51, my translation). Their principal function was to separate the infields from the outlying land.
According to Carlsson, the appearance of stone enclosures tells us about the increased significance of private ownership and the division of the infields into family units. There is no proof, however, that the properties had not been fenced before the stone enclosures appeared. We know that wood was used for fences in later times, and nothing contradicts that this was the case in earlier times. Together with the fact that the settlements from this earlier period are unknown, it is hard to prove that the stone enclosures indicate a division of the property into private ownership.
The greatest amount of research on stone enclosures has been done by human geographers, which might explain why the stone enclosures are primarily connected with agriculture and cattle breeding. This focus on an economic or functionalistic approach has in later times resulted in a request for alternative perspectives. Mats Burström (1994) has suggested some questions based on the assumption that stone enclosures, through their structuring of the landscape, created important places. He asks, for instance, whether the stone enclosures had a separation or a linkage significance and whether they were a source of social prestige. What he points out is "that the stone enclosures not only represent a fossil agrarian landscape but also a prehistoric world of ideas" (ibid 1994:70,my translation which the stone enclosures were not seen as a homogeneous phenomenon, it is possible to give rise to new ideas. An interpretation has to take into consideration functional as well as social and symbolic factors. These interpretations strengthen, rather than exclude, one another. If we look at the stone enclosures' function as fences, however, their effect could be questioned. In many cases they are only built of a single row of stones, which could hardly have kept the cattle away. If they really were fences they must have been reinforced with a wooden fence, something which has not been found in the excavations (e.g. Klindt-Jensen 1955:848).
Another important thing to note is the difference in definition of the stone enclosures. As pointed out, many of the authors treat the earth banks around square fields as stone enclosures. Here I follow Carlsson who treats the earth banks as cultivation marks from an earlier period, not as stone enclosures.

THE DATING OF THE STONE ENCLOSURES
The dating of stone enclosures is a complicated matter. Generally one has dated charcoal found in or below the stone enclosure. This could be criticized as an uncertain method because it is not clear what the charcoal represents. The charcoal under the stones only tells that it is older than the stone enclosures but not how much older. The different datings that have been obtained are from the early Iron Age to the Middle Ages (and even modern times). Bengt Windelhed, who has done local research in Vinarve, Rone parish, Gotland, believes that the stone enclosures are the result of building over a long time, and that the outlying land was gradually fenced (Windelhed 1984).Dan Carlsson, on the other hand, argues that the stone enclosures which are characterized by irregular lines in the landscape and with their stones sunken into the ground, almost certainly belong to the same time as the houses with stone foundations (around AD 200-5/ 600). "I do not know of any case where a stone enclosure as described above could be tied to any other period than the time with stone foundation houses" (Carlsson 1979:51, my  There are signs that point to the latter alternative, which means that the stone enclosures should not be treated as contemporaneous (Hyenstrand 1981).
My assumption in the following discussion is that the building of stone enclosures begins at the same time as the building of houses with a stone foundation, around AD 200. The stone enclosures were built during a long period of time, some even in modern time, but most of them are contemporaneous with the stone foundation house s. In thi s paper I will deal with the period AD 200-500.

THEORETICAL ASSUMPTIONS
How is it possible to discover some of the ideas that the prehistoric people had about the stone enclosures? Can one even assume that a common idea existed? I will not maintain that all groups in the society comprehended the stone enclosures in exactly the same way. Besides, it is likely that the meaning changed over time, when the enclosures were built in a different tradition and were used for other aims. However, I do think that we can find a similarity in the conception about the stone enclosures at a given time. This does not mean that every individual was fully conscious of and was able to articulate the stone enclosures' meaning, but rather that they were part of people's daily life and practice and their structuring of reality. Anthony Giddens distinguishes between practical and discursive consciousness, where the practical consciousness implies a knowledge of "how to go on" without the ability to put it in words (Giddens 1993:4). Pierre Bourdieu has also made a distinction between practice and theory. He thinks that rules and ideas about the reality are incorporated into the individual's body and that we can handle a situation practically without being able to describe explicitly how it is done (Bourdieu 1992:68 Another important subject to consider is the society's history. This means that it isn't enough to interpret how the stone enclosures have been understood but also interpret the motives behind their origin. What needs and strategies lay behind the building of stone enclosures? Bourdieu points out the importance of seeing the history in an analysis of a society. The field where people act as well as the habitus is a product of its history. "The habitusembodied history, internalized as a second nature and so forgotten as history is the active presence of the whole past of which it is the product. " (Bourdieu 1992:56). I will therefore also consider the historical background of the stone enclosures.

ARGUMENTATION
In an earlier paper on houses with stone foundations I have suggested that stone as building material is an important startingpoint when interpreting stone enclosures and houses (Cassel 1994). The rights to or ownership of the land was in earlier times seen as a "natural" source of power. When this changed and other ways to gain prestige became possible, the landowning elite felt a need to legitimate their power. This was partly done by building the houses and enclosures in stone and in this way emphasizing the farm's importance for the social order. Stone is a durable material and had also been used to mark the graves. By building houses and enclosures with the same material as the graves, one used the symbolic significance of stone in a new context. In addition, the stone enclosures' widespread presence in the landscape made them an appropriate medium for argumentation. One question, however, is whether the stone enclosures had other functions than keeping the animals away from the infields. If we look at the stone enclosures as mere fences we will see in them a separation function; they demarcate the infield from the outlying land and separate humans from animals. But one can also see the stone enclosures as linking together farms and graves. Some authors have paid attention to how the enclosures tie farms together in a village organization or "enclosure society" (Sw. hägnadslag) (Fallgren 1993;Widgren 1983).In this paper I do not intend to discuss whether or not the farms which are linked together by stone enclosures can be called a village, but to discuss why people felt it necessary to link the farms together.
What did they tie together with the stone enclosures, and how did these lines in the landscape, which linked farms and graves together, affect the people? I would like to see the stone enclosures as an argument in the structuring and organizing of the society.

The argumentation
was supported by their duration and presence throughout the landscape. The fact that the stone enclosures guided people's movements in the landscape Carrent Swedish Archaeology, Voh 4, 1996 towards the graves, indicates how one used the history and the ancestors in the structuring of society. The stone enclosures were linking together ancient times with present time, and future time. But what did the farms that were tied together by the stone enclosures have in common? Before we try to answer this question we need to take a closer look at the society's organization.

THE ORGANIZATION OF SOCIETY
The society in the early Iron Age is usually described as a chiefdom. This term is used for many different societies and does not tell us more than that they were stratified. It is not my intention here to define the early Iron Age chiefdom, but I will try to point to otze important principle of organization, namely, kinship. In his work on the early Iron Age in Norway, Knut Odner describes the society as a redistributive chiefdom without a developed market. Relations with kin were established over large areas, which made it possible to exchange prestige-goods (Odner 1973).
According to Lotte Hedeager, the family was most important in the Danish society for keeping the power within a given group. In the early Iron Age the society was organized around the kin: "It also follows from this that in tribal society, that is, in a chiefdom, both production and reproduction are organized through kinship groups" (Hedeager 1992:87). Heiko Steuer has also shown in his study of the German societies, that the base in the society was the extended family ("familia"). Usually several extended families were grouped together in kinship groups and shared, for instance, a common handicraft (Steuer 1982). Bourdieu has noted that the family (extended family or kin) in the traditional Algerian society was a model for how the other social structures in society were constructed (Broady l 990:262). In line with this, my point of departure for the following discussion is that the society on Gotland in the early Iron Age was organized around the family and kin. This applies particularly to the rights to the land, which was kept in the family and was not for sale. Larger, collective farms (so-called primary units) were divided into family units and were moved to the border between infields and outlying land. (The outlying land was not divided on separate farms. ) (Carlsson 1979:85f) Such a movement of the settlement and division of the land requires, as I see it, some kind of centralized power (cf. the regulations (Sw. slciften) of the villages in the 18th and 19th centuries). These prerequisites did not exist on Gotland in the Roman Iron Age, and such model» are therefore not appropriate when discussing the early Iron Age. I do believe, on the other hand, that an analysis of the systems of enclosures could tell us something about the social structure which affected the organization of the settlement. Fig. 1 shows a relatively clearly defined area in Buttle parish, Gotland. There were remains of at least ten stone foundation houses, two of which are gone today and another i» badly damaged. These ten houses should represent at least four farms. There are also remains of stone enclosures which partly encircle an area and partly cut right through this area. Lying central along this latter stone enclosure are three graves. In the area there are also two wells (Sw. bryor) and heaps of fire-cracked stones. East of the infields are 64 so-called grinding furrows (Sw. svcircisliptzingsrctntzor).

TWO AREAS WITH STONE ENCLOSURES
The stone enclosures which stretch through the area diverge in the middle and form two arms. Both of these arms lead towards gravestwo large cairns and one small stone-setting. The graves are probably from the Bronze Age. My opinion is that the stone enclosure was built for the purpose of linking together farms with the graves, the living with their ancestors, the present with the past. What, then, do the farms along the stone enclosures have in common? My suggestion is that they represent families who create a kinship group. Enclosures and grave paths guide the way through the landscape and connect the farms in a "natural" way. The landscape's organization and structure corresponds to the social structure, with the family as a base. In a society like this, one could expect other common fields of activity besides the building of enclosures, for instance, some handicraft as Steuer has suggested (see above). Since there have been no archaeological investigations in the area it is impossible to show any specialization, but the number of houses on each farm could perhaps give an indication of where we should look. Three of the farms consist of two houses, while the one to the west consists of four. Perhaps it is here we can find some shared activities which tied the community even closer together. The other example is from Sjonhem and Vänge parish. Gotland (Fig. 2).This picture is more complicated, and it is not clear how the area is divided into infield/outlying land for instance. The stone enclosures extend in every direction, and the possibility of an interpretation is even more complicated owing to cultivation in later times. Some excavations have been done in the area, for instance, a trench through a house with a stone foundation and through a stone enclosure in the middle of the area. One C14-dating from the house gave the result AD 230+90 (Carlsson 1979:102 (Cassel 1994) I have shown that the amount of precious goods, such as Roman imports and gold, increases in the burial s in the late Roman Iron Age. My interpretation is that, from now on, it became possible for a larger part of the society to appropriate such goods. Maybe this was made possible by the the fact that some men did military service on the Continent or by an increased "trade". In earlier times only the landowners could produce the surplus that made it possible to appropriate prestige-goods. By the late Roman Iron Age the situation had changed, and the landowning elite did not have this opportunity to legitimate their power. According to Pierre Bourdieu, the dominance in a pre-capitalistic economy could only be maintained through direct and personal contact, since the society did not have any institutional power. The dominance is maintained and legitimated through symbolic capital, which does not need to have any "real" value. "In such a context, the accumulation of material wealth is simply one means among others of accumulating symbolic powerthe power to secure recognition of power" (Bourdieu 1992:131).The objects which in the early Roman Iron Age had functioned as symbolic capital, were no longer exclusive for the elite, and new strategies were necessary for legitimating power. In this context they chose to emphasize the farm and the land, which in the previous period had been the source of surplus production and domination. By building the houses and the enclosures of stone, they tried to "affect" the times to come: the stone would guarantee an unchanged future.
What about the stone enclosures and kinship system? Lotte Hedeager tells us that great differences occur in the late Roman Iron Age in Denmark. The family, which had been the base and the central point in society, was replaced by a centralized power with professional warriors: "The last ancient tribal traditions have gone" (Hedeager 1992: 246). Did this affect the people on Gotland, and was the old kinship system challenged by a new organization based on other loyalties than the family? Maybe this was why they felt it necessary to emphasize and embody the family organization through the stone enclosures. When people moved around in the landscape, they "walked" the social structure. When the stone enclosures and the houses with a stone foundation came to an end in the Migration period, it could be a sign that this sort of argumentation wasn't needed any longer. Maybe by then another form of organization had replaced the kinship system.

SUMMARY
In this paper I have tried to show that dif-Cttrrettt .sii eilis!!, 4rel iteclo"-s', Vol. 4. !996 ferent perspectives such as historical context, functional aspects and analyses of the everyday practice do not necessarily exclude one another or further confuse the picture of prehistory. Different perspectives give us instead the possibility to understand how complex the reality is and they stimulate new interpretations of the past. This does not have to be done by rejecting the prevailing ideas, such as that the stone enclosures divide the infields from the outlying land, but by approaching some other ideas about the enclosures. I have argued that the stone enclosures which link together farms and ancestors' graves were an expression of how the society was organized around the family and kin. The stone enclosures are always pre-sent in the landscape and when people move along them they structure their reality. The fact that they are part of people's everyday practice means that they have a reproductive and preserving effect. The family structure was the base for the organization of society, and the stone enclosure embodied and naturalised this condition. The historical reasons for the stone enclosures could originate in the fact that the society was changing. The kinship structure was challenged and the stone enclosures were built as an argument against change. The argument was strengthened by their linking of both time and space.