Kinship , Marriage and Traces of Social Interaction Aspects on the hunter-gatherer societies in northern Sweden during the Bronze Age and early Iron

A central problem in the explanation and description of decentralized hunter-gatherer societies is what kind of traces different sorts of social interaction will leave in the archaeological material. It is here suggested that the distribution of asbestos-tempered ceramic ware in northern Sweden during the Bronze Age and early Iron Age was a result of the social interaction between different groups. The decoration on the cermic ware is suggested to represent traces of intermarriage relations. It is further argued that the hunter-gatherer societies were open to interaction with other groups and individuals.

This article will focus upon two features that have been important to the explanation and description of decentralized hunter-gatherer societies: i) the existence of isolated and more or less "closed" societies, and ii) societies "open" to interaction with other huntergatherer groups and with an extended mobility in respect to social contacts (that i», mobility and interaction between neighbouring groups producing and reproducing social relations). A central question connected with this issue is what kind of traces different sorts of interaction will leave in the archaeological material. The article aims to illuminate a more specific aspect of the distribution of asbestos-tempered ceramic ware in Norrland and its relevance for defining kinship relations and interaction between groups. I will argue that the acceptance and distribution of asbestos-tempered pottery should be explained by interactional strategies and intermarriage relations between different groups, who were already linked together by a network of previous exchange and relations. The exchange of marriage partners was a guarantee for further relations between different groups. The pottery was partly involved in manifesting affinity and intermarriage relations between the gl'oups.
The geographical setting is northern Sweden (Norrland), roughly defined as north of the river Dalälven. The object of interpretation is the hunter-gatherer societies in Norrland during the Bronze Age and early Iron Age, that is, between 1800 B.C. and the centuries around the birth of Christ. The Hoting-area in Tåsjö parish, Ångermanland is chosen as an example to represent the northern hunter-gatherer societies in general.  (Forsberg 1985:274). He suggests that the settlements during the Bronze Age was organized as a mobile system between two areas within the territory, the forest area and the mountain foothill area. The movement was seasonal and the base camp localisation during the winter, spring and early summer in the forest area was, according to Forsberg, determined by the main economic activities such as fishing and elk hunting, etc. A number of exploitation sites were located around the base camp for different activities. During the rest of the year the whole hunter-gatherer group moved to another base camp in the mountain foothill area, where the main activities were the hunting of reindeer and the procurement of raw material for the making of tools and points. Between the two areas in the seasonal movement cycle there was a transient camp. Although Forsberg points out a strict descriptive purpose for the model (1985:275) it is not clear whether this system was regularly or irregularly "open" to contacts with individuals or groups from outside. In a later paper Forsberg suggests that the rock carvings at Nämforsen and Norrfors in Angermanland and Västerbotten were aggregation camps for two separate hunter-gatherer systems or societies. Nämforsen is suggested to be an aggregation camp for the hunter-gatherer groups along the Ångerman river, and Norrfors is suggested as a similar camp for the hunter-gatherers at the mouth of the Ume river (Forsberg 1990).According to Forsberg (1988) and Spång (1991), it is the rich resources of salmon during the summer that are of primary interest for the location of an aggregation camp. This assertion, however, explains nothing about the meaning of the term "aggregation camp" and the presumed social and ritual activities that took place there. I think there also has to be something less functional in the explanation than good resources for fishing, if we want to get a deeper explanation and understanding of the rock-carving sites as aggregation camps. A more adequate understanding is perhaps to be found in the connection between rockcarvings and the contact with flowing water, and the associated ritual symbolism and meaning that was related to these specific "close to water" activities. From the variations in the site materialhigh or low diversity of scrapers and points, etc. within different sites -Forsberg comes to the conclusion that the variations are due to differences in site utilisation and subsistence strategies between the two areas in the territory (1985,1988). This perspective on hunter-gatherers adaption to a set ecological system is in my opinion limited because it doesn't tell us much more about these people than about a group of animals moving to the next waterhole. Hodder has suggested that each particular historical context must be studied as a unique combination of general principles of meaning and symbolism, negotiated and manipulated in specific ways. The archaeological data do not simply reflect human behaviour in a predictable way. It is constituted and transformed in different social strategies. It is therefore important to integrate adaptive variability with an interest in culture and meaning (see Hodder 1992).
What, then, is the cultural and symbolic meaning of the site variability in the inland of Norrland? What can the variations in the site material tell us about the social strategies of the hunter-gatherer groups? These questions are very important and I will return to them later.
If we for a moment accept the second presupposition (ii) mentioned above, about societies "open" to interaction and contact between groups, it is important to start with the question of why we would expect to find places for social interaction and contact and what they might look like. Later on in this paper I will argue that the existence of iso-Curren(Swedish Archaeology, Vol. 4, l996 lated and socially "closed" systems of hunter-gatherer groups is for several reasons not plausible in the interior of Norrland. First of all, why are rock-carving areas such as Stor Norrfors, Nämforsen, Glösa, Gärde and also the Annsjö-area in the interior of middle Norrland, the only type of sites to consider when trying to reconstruct a social network with nodes of contact between decentralized hunter-gatherer groups? Is it warranted to exclude the existence of a network, with meeting-places or nodes of contact, when not finding any complex rock carving areas? I do not think so and I do not find that Forsberg's model is sufficient in respect to these questions. My opinion is that we cannot convincingly explain any settlement pattern with a strict ecological adaptive model. More often it tums out to be purely circular reasoning. The distribution of prehistoric settlements can always, in one way or another, easil y be explained by favourable ecological conditions simply because of the presence of certain animal species, for example: l) reindeer-hunting in the mountain foothill area, 2) elk-hunting and fishing in the forest-river area, 3) salmon-fishing along the rivers during the summer, etc. Another reason for not accepting this eco-economical approach is that it is too simplistic and, as Karl Polanyi and others have held, distribution and exchangelike other economic relations is "embedded" in social relations and therefore a matter of greater complexity (Bloch 1975:40). investment with "delayed returns". This sort of manipulation of the other part does not necessarily imply anything directly negative in the everyday relations. If I give you a gift we are both clear about you giving something else back some other time. Bourdieu argues that the time factor is important in the practice 'of giving gifts and counter-gifts.

DIFFERENT SORTS OF INTERACTION -DIFFERENT TRACES
Interaction has in this context a very wide meaning, from exchange and distribution to more complex mutual processes such as gift exchange and marriage relations. Manipulation of the other part, for example by gift exchange, is of vital importance when creating social dependencies and (long lasting) social relations. The giving of gifts and counter-gifts involves a strong element of social  (Forsberg 1985). "Society" should here be understood as a group of "families" or lineage groups sporadically or continually acting together very closely, occupying a territory and simultaneously producing/reproducing a common social structure. It is not obvious that different societies can be detected and defined through the archaeological material. The wide geographical distribution of similar material expressions, asbestos-tempered ceramic ware for example, that occur in almost all of Norrland (Hulthén 1991), shows very clearly the existence of a developed social network with contacts over large areas. At the same time do these structural similarities (or material culture in general) have a tendency to hide territorial or other social differences, mainly because of human manipulation and transformation of the material in different strategies (see Burström 1991).The site material from excavated settlements in the interior of Norrland is often chronologically mixed and therefore not very easy to classify. There is often evidence for long continuity of occupation. An important question, mentioned above, is what kind of lasting traces different sorts of interaction could leave on a specific site.
Exchange networks and the performance of religious rites are two aspects that are often mentioned when interpreting extraordinary variations in the material. Other aspects of contacts between individuals and groups may not be directly visible or empirically evident at all, such as exogamic marriage relations, the giving of gifts and counter-gifts, competition and transmission of innovation, to name a few. These relational aspects are rather to be detected through the process of interpretation within a framework of social theory.
The rock carvings at Nämforsen in Angermanland, located along the Ångerman river, are mainly interpreted as a site of great ritual importance. Tilley has suggested that it was an important site for the hunter-gatherer people of the interior in their contacts with the farmers along the coast. He places Nämforsen in a context of unequal and exploitative exchange relations between the metal-producing farmers and inland huntergatherers. According to Tilley, the rock carvings result from increasing exchange relations between the inland groups and the coastal population (Tilley 1991:155-164).
Despite the unique and extraordinary character of the area, the rich and heterogeneous find material from the excavated settlement at Nämforsen (flint, asbestos-tempered ceramic ware, seal bones etc. , according to Baudou 1990) is likely to be explained by interactions between people over both long and short distances. It is not reasonable to believe that the activities at Nämforsen only had significance for just one society or social group.

ASBESTOS-TEMPERED CERAMIC WARE
Asbestos-tempered ceramic ware (ATCW from now on) occurs all over northern Scandinavia, Finland and across the border into north-western Russia. The ATCW in northern Scandinavia has been subjected to Current Swedish A rehaeology, Vol. 4, l 996 several analyses and grouped and classified to variation and geographical distribution. ATCW is generally suggested to appear during the Bronze Age and disappears a few centuries A. D. Hulthén has divided the asbestos-tempered ceramic material into two groups, asbestos ware and asbestos-tempered pottery, depending on how much asbestos is used as temper material. Hulthén suggests that the asbestos-tempered pottery was connected with early bronze casting and metalworking in Norrland. The main argument is that asbestos minerals have one important property in common, namely high resistance to heat (Hulthén 1991).Olsen and Jörgensen (1988) suggests that ATCW was involved in interactional contexts where the interior hunter-gatherer population used the ATCW in social strategies, expressing group solidarity. Those strategies were a response to an increasing difference in the subsistence eco- The use of the archaeological material as evidence for defining ethnic groups and boundaries has also been a very attractive subject. In the recent archaeological debate it has been argued that it is relevant to connect specific archaeological data within a defined area with a present ethnic group (the Saami people), occupying the same area (Baudou 1987:9;Zachrisson 1984 tuting the cultural features which at any time distinguish an ethnic group because these overt cultural forms are determined by ecology as well as by transmitted cultural and social components. "We must expect to find that one ethnic group, spread over a territory with varying ecological circumstances, will exhibit regional diversities of overt institutionalized behaviour which do not reflect differences in cultural orientation. How should they then be classified if overt institutional forms are diagnostic?" (Barth 1969:12).Only some of the overt cultural forms are used to signal ethnic identity. Ethnic groups, according to Barth, are seen as a form of social organisation, and primary emphasis should be given to the fact that ethnic groups are categories of ascription and identification by the actors themselves (Barth 1969). Despite the methodological problems concerning the possibility of distinguishing ethnic groups in the archaeological material, there is good reason to be pessimistic about how to give a meaningful definition of the word "ethnicity". Barth's critique of the traditional view of defining ethnicity stresses only one of many perspectives. I don't support the above arguments for connecting the asbestos-tempered ceramic material with one ethnic group, historically known as the Saami people. I cannot see any plausible reasons for this ethnic connection.
Several excavations of settlements along the coast in Norrland has shown finds of ATCW (Fig. 3 in Hulthén 1991r47 shows a few of these finds). The find situation speaks against the suggested ethnic division between the Saamis in the interior of Norrland and the "Scandinavian farmers" along the coast.
Odner has also criticized the correlation between asbestos and the Saamis, suggested by Baudou and Olsen, from a linguistic point of view (Odner 1983(Odner , 1992). I will return to this issue and put the ATCW in a context of social interaction between different groups in Norrland.

KINSHIP RELATIONS
The meaning of the term "peer polity interaction" (PPI) is, according to Renfrew, an autonomous socio-political unit which doesn't need to display any notably developed or differential system of government or of administration: it is sufficient that there exist procedures for decision-making which habitually work and in practice modify or affect the members behaviour. A polity is not subject to the jurisdiction of any higher power. Autonomous socio-political units do not generally exist in isolation, but have neighbours which are analogous (peer) in scale to them (Fig. 2). This definition may very well apply to the picture of the north Scandinavian hunter-gatherer societies. An important question according to the concept of PPI is why we should find structural homologies and similar elements repeated throughout a region. One explanation for shared elements or structural homologies within a region comes, according to Renfrew, from the interaction between peer politiespeer polity interaction (Renfrew 1986:1-10). If we take ATCW, for example, which is a widespread element (structural homologies) throughout northern Scandinavia, the concept of PPI could be useful when we want to explain its widespread acceptance in a region. Acceptance of a new element (an invention) in one society may facilitate or sanction it within another (Ibid 1986:10).
Although arguing the importance of the The ATCW in Norrland mainly appears in domestic contexts, which doesn't tell us very much about its relevance for male or female practices. According to Hulthén, the most important functional property of asbestos is heat resistance. Asbestos-tempered ceramic ware may well have been invol ved in the introduction of the bronze casting technology (Hulthén 1991),but that was probably one of many functions. It is reasonable to believe that ATCW also had other, practical and symbolical significance (Linder 1972).
The ATCW is not necessarily only to be connected with special activities like bronze casting. There are too few moulds, crucibles and bronze objects found to support the picture of any large-scale production of bronze artefacts. The rather frequently distributed ATCW may also have been involved in ordinary domestic practices, such as heat keeping containers for different cooking activities, and at the same time has also expressed family and kinship relations through the decorations and ornaments on the vessel body. The wide distribution of sim- Fig. 2. Peer polityinteraction. Strong interaction between autonomous socio-political units within the region is of greater significance than external links wi th other areas. From Renfrew 1986. '7. ilar decoration categories is here suggested to be connected with intermarriage and kinship relations (Fig. 3). The different decoration categories may represent different alliancesystems over time. A fine-scale chronology of the ATCW is however not possible to present. Anyway, a system of alliances should be understood as a system whereby descent groups or other kin groups are linked by a rule of prescriptive or recurrent marriage so that the groups remain in a affinal relationship to one another across generations (Keesing 1981:507).
If we look at the distribution of decoration categories (Fig. 3), the vast distribution of ATCW is likely to be explained by the following hypothesis: The acceptance of asbestos-tempered ceramic ware in one huntergatherer society, which seems to occur roughly at the same time as the introduction of the bronze casting technology, facilitated and sanctioned its acceptance in the neighbouring group. This was made possible by the earlier established "ties" or relations between neighbouring groups. These "ties" were once created by kinship and marriage relations and re-established across genera- Fig. 3. Distribution of different decoration categories within Hulthin's investigated area in Norrland. The three hypothetical "districts" contain various compositions of decoratir&n categories, and sites. The main interest of this figure is the distribution of similar decoration categories, not the hypothetical "districts" (Hulthén &991&47  tions. Different decoration categories on the vessels are thought to represent different alliances and kinship groups. The innovation of ATCW was probably not only involved in bronze casting activities, as suggested by Hulthén, but was also widely distributed in more ordinary and domestic practices, perhaps as heat-keeping containers when cooking food. In these domestic contexts it must have been important to signal social position and kinship affiliation through the symbolic pattern that was decorated on the pottery. The decorations seem probably to have been involved in the hunter-gatherer people's classification of the social world in order to understand it as well as to reproduce the established kinship structure. It is important to stress that it was not the vessels that were exchanged between different groups. It was people involved in intermarriage relations who were bringing the tradition along with them to where they chose to live. Similar decorations on the pottery were the result of these intermarriage strategies.

TRACES OF INTERACTION
The Hoting-area, Tasjö parish, in north-western Angermanland, is centrally situated in the interior of Norrland. The area consists of about 80 prehistoric settlements or activitysites dating from the Mesolithic to the Iron Age, all sites located around the lake Hotingsjön (Fig. 4). It is regarded as one of the densest areas in Norrland concerning prehistoric sites.
The Hoting lake constitutes a part of the Fjällsjö river water system which, further downstream flows out in the Ångerman river.
The number of excavated sites in the Hotingarea is about 13. Asbestos-tempered ceramic ware is found on at least eight sites around the lake (SHM and ATA). The distribution of sites in the Hoting-area containing ATCW is mainly concentrated to two areas, one easterly and one westerly, except for one site in between. There is a close connection between ATCW and subterranean huts (Sw. skärv-stensvallar), which are suggested to be base camps (Lundberg 1986). One can also notice that nearly all of the sites with ATCW are located on islets and islands. The rest of the sites are located along the lake and river banks. It is also important to point out that the distribution of ATCW and its geographical representation is closely connected with the excavation situation in the area. are due to differences in chronology we still need to explain why these various decoration patterns are so widespread. I find no reason to accept a mobile society in the middle Norrland area with the same groups leaving pottery all over. I rather suggest that relations and contacts existed between semisedentary social groups (that is, base camp settlements with huntingand activity-sites) mainly living in one area during the whole year round. If we accept thi» picture we could, according to the theory of marriage and kinship relations presented above, further discuss the ATCW s relevance for alliance systems and intermarriage relations between groups. In accordance with the distribution map (Fig. 4)  and outsiders. Maybe the exchange of marriage partners was a guarantee for further transactions and economic relations and hence the creation of new bonds between families and groups. According to Bourdieu, marriage is almost always set up between families already linked by a whole network of previous exchange, underwriting the specific new agreement (Bourdieu 1990:H5).
Maybe the decorations on the asbestostempered ceramic ware were made by men or women who had been exchanged to another "village" or settlement area, sometimes very far away from their own lineage group. Similar decorations are suggested to signal intermarriage between different groups. Perhaps men and women in intermarriage relations constituted a keyposition in ensuring further economical transactions as well as social relations established between groups. Similar decorations on the pottery existed because of important marriage relations, a guarantee for maintaining further relations and it was produced by married men or women in their new residence.

CONCLUSION
The purpose of this article has been to stress the importance of interaction between different hunter-gatherer groups but also to question the existence of socially "isolated" systems in Bronze Age and early Iron Age Norrland. In previous archaeological studies the focus has been on extraordinary features such as the rock-carving areas of Norrfors and Nämforsen when discussing places for aggregation and contact. From my point of view, this approach tends to obscure other possibilities of explaining the archaeological material in terms of social strategies and interaction between groups. It is most relevant to emphasize the character of different social relations between groups when discussing the structure of decentralized hunter-gatherer societies.
In this article asbestos-tempered ceramic ware is studied for the purpose of illumina-ting its relevance for social relations between different groups. Asbestos-tempered ceramic ware is widely distributed over northern Scandinavia and it has sometimes been related to the ethnic process of the Saami people. It has been suggested that this ethnic process was the result of contacts with Scandinavian farmers along the coast (Baudou 1987; Olsen k Jörgensen 1988).Against this hypothesis is the increasing number of sites containing ATCW along the coast. There is no major difference between coast and interior in respect to the distribution of ATCW. The question of defining ethnic groups in the archaeological material i» also connected with a number of difficulties. Ethnicity is not just a passive reflection of the economic base of subsistence. According to Barth: "It is important to recognize that although ethnic categories take cultural differences into account, we can assume no simple one-to-one relationship between ethnic units and cultural similarities and differences. The features that are taken into account are not the sum of "objective" differences, but only those that the actors themselves regard as significant" (Barth 1969:14).
I have tried to make two interpretations of what the ATCW in northern Scandinavia represented, one high-level aspect on the general distribution of ATCW and one more concrete view on the particular context around the Hoting-area. The aspects put forth in this paper put more emphasis on the social relations than on ecological adaption in the hunter-gatherer societies. The wide acceptance of ATCW in northern Scandinavia is here partly explained by the concept of peer polity interaction. The distribution of ATCW was not the result of trade or exchange but was accepted as an idea. An autonomous sociopolitical unit (polity), the Hoting-area for instance, did not exist in isolation but had neighbours that were equal (peer) in scale.
Acceptance of a new element (ATCW for example) in one polity is therefore likely to facilitate or sanction it within another (Ren-frew 1986). Distant marriages, which result from elaborate strategies and are expected to yield alliances, are, according to Bourdieu, a kind of longand short-term investment designed to maintain or increase symbolic or relational capital (1990).Interactional activities of various kinds between groups are hence suggested to be closely connected with marriage relations and kinship. Intermarriage and exogamic relations make it possible to transform outsiders into insiders by means of affinity. Blood bands are among the strongest relations which make it possible to produce and reproduce social relations with people outside the community and create social dependencies. The distribution of asbestostempered ceramic ware is thought to have resulted from such interactional strategies outlined above, perhaps also manifesting affinity, alliance and solidarity with other groups. Similar decorations on the vessels are here suggested to be connected with intermarriage between groups. The traces of interaction between different hunter-gather groups do not necessarily have to be extraordinary in nature.
English reui sed by Laura Wrang. ACKNOWLEDGMENT I would like to thank Bjirrn Eriksson, Depart&nent of Phi Josophy, for crrnstructive comments on this a rti cle.