On Studies of Task Bifferentiation between Men and Wotnen in the Scandinavian Iron Age

Reiterated and cursorily criticised generalisations of attributes for male and female in grave goods, have since the first half of the nineteenth century created an oversimplified yet politically intricate image of a specific task differentiation between men and women in prehistory. Ideals of male and female roles and tasks in the interpreter's contemporary society have been described as universals in terms of binary oppositional pairs, or spheres, such as private/domestic-public. The dichotomies used for analysing and attributing male and female tasks have given preference to stereotypes, and the very formulation of the oppositional concepts for activity areas expresses ideological valuations ofmale and female. This article stresses the need for analysing the origin of concepts, and it seeks new and alternative ways of perceiving task differentiation.

specific task differentiation between men and women in prehistory.
Ideals of male and female roles and tasks in the interpreter's contemporary society have been described as universals in terms of binary oppositional pairs, or spheres, such as private/domestic-public. The dichotomies used for analysing and attributing male and female tasks have given preference to stereotypes, and the very formulation of the oppositional concepts for activity areas expresses ideological valuations ofmale and female. This article stresses the need for analysing the origin of concepts, and it seeks new and alternative ways of perceiving task differentiation. Louise Sttöbecl; lnstitute of Atc/taeolo~v, Lund Utt/vetsitv, Sandgata» l, SE-223 SO Lund, Su:eden. The very first time sex was mentioned as a concept in studies of prehistory was in 1837, in Mecklenburgisches Jahvbuch. In a programme with instructions for how surveys of prehistoric graves should be conducted, the authors expressed a desire that female and male artefacts found in the graves should be the point of departure for sex determinations of the deceased. The sword was classified as a typical male artefact, and especially the sewing needle was interpreted as a characteristic female artefact (Hjorungdal 1994(Hjorungdal :67, 1997.The weapon andjewellerypair was a second common material criterion of male and female in archaeological interpretations during the nineteenth century (Hjorungdal 1997:38 She notes that in Sogn, western Norway, women have been buried with both female artefacts, like textile implements, and male tools like axes, files and arrowheads. Dommasnes postulates that women buried with male tools have taken over tasks which were formerly performed by men. The reason women carried out tasks that traditionally were not regarded as appropriate for women was, according to Dommasnes, that men had left the farm for trade, piracy and warfare in foreign territories. Imported goods in Sogn indicate that the region had contacts with foreign territories, and the author refers to written sources when she says that it was the men who travelled long distances (Dommasnes 1987:65ff). When men's time and effort were spent elsewhere than at the farm, especially during the Viking Age, the women left at home apparently took responsibility for the farming and consequently the women also took over certain male attributes.
Dommasnes says that it is even possible to make statements about the status of men and women in the Late Iron Age in Norway, on the basis ofthe ratio ofmale/female graves. It is regarded as indicative of a high ranking individual in the grave, if the deceased is buried in a grave that is visible for archaeologists through grave furnishings or markers above ground (Dommasnes 1987:69, 71).

Dommasnes suggestsin accordance with
her study of the ratio of male/female gravesthat more women obtained high rank in the ninth century than in any other period, and that the number of high ranking women was greater in the coastal areas than in the interior during all centuries from the 7'" to the 10'" (Dommasnes 1987:71'.The interpretation is consequently that women stepped in and took over typical male roles at the farm during the men's absence, and that this rendered the women a higher social status (Dommasnes 1987:76).If this analysis was meant not only to attribute the Late Iron Age woman new tasks, but also to make her an active person in the development of the society, then the interpreter has missed the fact that women were still described as lacking influence and power in society and history. Arwill-Nordbladh 1994:36).These oppositional categories have been used to analyse and illustrate differences between men and women. But as I stated earlier, the dichotomous concepts imply power relations through their construction as pairs and through their definitions. In categories such as private/ domestic-public and nature-culture, the terms asymmetry and hierarchy are integrated into each other, and the oppositional poles can perhaps be said to be explained by each other.
The concept of asymmetry, however, originally denotes horizontal differences between individuals or among groups of inhabitants, while hierarchy describes vertical difierences, that is, power relations and inequalities. Asymmetry and hierarchy may be linked in reality, but the two phenomena probably appear as a pair to a greater, or to a lesser, extent in certain societies than they do in other societies. It is also likely that the co-variat!on between horizontal and vertical differences among prehistoric inhabitants finds widely different expressions in various cultures (Thurén 1996:74 categories as in previous research will, however, limit the alternative interpretations of the design of gender relations (Prestvold 1996:24).The generalisations and the stereotyped categorization will still characterise these alternative pictures, and the background of the differentiation into two spheres will remain unproblematized and will not be modified by this kind ofreassociation ofmale and female.

DECONSTRUCTING SPHERES, AND CREATING RELATIONS OF CO-OPERATION
There is more than one way of analysing and describing past cultural or societal organisations in terms that do not presuppose hierarchy and prescribe subordination. There are, however, at least two essential items in research connected to each other, which should be regarded when studying male and female activities, and especially when the aim is to create new and alternative ways of describing task differentiation between men and women in prehistory. I have already stated that the point of departure for studying male and female activities should be a critical evaluation of different analytical concepts' inherited implications of power relations. This evaluation should be completed with an analysis of the very tasks men and women actually are likely to have performed in accordance with the source material, and what the interpreter's association between artefact and gender role or task is based on. The two moments emphasise that images, attributes and activities associated with men or women in a given context have to be investigated, not assumed. This corresponds to the definition of gender as a socially and culturally dependent variable, possible to express in several different images.
One way of avoiding projection of the contemporary value-loaded sex role pattern onto prehistoric societies is to operationalize questions of male and female by studying many activity areas. Instead of forcing the various activities of any society into two diametrical spheres, it would be possible and probably even fruitful to study more than a better with a multi-faceted model for task differentiation, than with a "two-sphere" model that implicitly identifies activities with sex roles. Making shelter for cover and protection could be a potential alternative activity area or category in a new kind of research on task differentiation. (The term "shelter" was proposed by Dr. Elisabeth Iregren, in a discussion. ) Paying attention to activities such as making shelter, brings together various tasks which have been interpreted as either male or female in previous research. Making shelter can include tasks like sewing clothes and building houses. Both tasks are performed in a sphere that could be called "at home"; not domestic perhaps, but at least at home. It would, however, still be easy to give way to the traditional dichotomous thinking about the sexes in a study like that, if the study of making shelter solely dealt with the task of running a needle through pieces of cloth or working with an axe on timber. There is consequently a need to develop the suggested perspective further, and this brings us to a critical analysis ofthe definitions ofactivities.
Analyses of the definitions of already gender-attributed tasks in archaeological research have shown how former definitions of activities have been restricted to the performance of one type of work within an activity which in fact consists of many tasks (Wright 1991:198ff;Stig Serensen 1996:Slff). A survey of what the types or sequences of work are that have been interpreted as crucial to the performance of an activity, will presumably show that it is not just the preserved artefacts but also the interpreters' images of how activities are constituted, that have made the archaeologists regard them as significant. Hunting, farming/ fieldwork and metalwork are mostly interpreted as male activities in traditional research, and the three have been identified with tasks within these activities that focus on the types of work that are associated with a male agent. Hunting has come to be identical with the killing of game, fieldwork during the Iron Age is synonymous with steering the plough from the rear, and metalwork is described as casting, forging and smithing the metal (Olsen Bruhns 1991:427;Stig S@rensen 1996:Slff). These more or less explicit definitions of hunting, fieldwork and metalwork probably derive from the interpreters' associations between killing, steering and hammering, and male courage, endurance and physical strength. It should be noted that these activities have been interpreted as characteristic of Scandinavian prehistory, according to the designations of prehistoric epochs: the Stone Age (the Hunter Stone Age, and the Farmer Stone Age), the Bronze Age and the Iron Age. If the other tasks necessary for the performance of the postulated crucial tasks in the activities mentioned above should be regarded, hunting would be analysed as Louise Ströhech an activity including tasks such as looking and spying for game to hunt for meat supply and/or for their supply of fur. The possible calling of game with characteristic sounds or cries, and the strategic driving of the chosen game towards an abyss or in the direction of the persons who intend to kill the animal(s), are tasks that also have to be conducted if there is going to be a successful hunt at all. Leading the oxen to the field, hitching them up and walking in front of them, guiding the animals if needed, and sowing, are all tasks that are essential when fieldwork and ploughing are carried out. besides the steering of the plough from the rear (Olsen Bruhns 1991:427).In addition to casting and forging, metalwork involves tasks that are connected with prospecting, producing firewood and charcoal, and shaping and decorating the objects (Stig Serensen 1996:56).If we return to the example of the making of shelter, the redefinition of activities I have argued for would mean that sewing clothes should include the making of cloth for a dress. The activity ofbuilding houses should on the other hand include tasks such as collecting and obtaining the different materials required for building walls, roof, etc. The building materials from different environments have also been handled with different techniques, which rely on various kinds of knowledge, skill and experience.
A redefinition of several activities is called for in order to do studies of the relations between men and women, and not between male and female spheres, in prehistory. Studies in which activities are analysed as being composed of sequences of tasks or concurrently executed tasks dependent on each other, break up the stereotyped dichotomous categorization of male work contra female work. They pave the way for gender studies of task differentiation which notice age-linked sexual differentiation, family group activity, co-operation between men and women, and the existence of non-gendered tasks and activities (Olsen Bruhns 1991:427;Wright 1991:198ff).But doesn't this kind of analysis eliminate the grounds for studying task differentiation between men and women in prehistory? The answer is that it will only blur the often simplified picture of a rigid differentiation of work and activities that has been presented in most archaeological texts. lt will wipe out that picture's great resemblance to the Victorian bourgeois gender ideology, which lingers on in societies in the West.

ELUCIDATING PREVIOUSLY FOR-GOTTEN AND NON-GENDERED WORK
Although much of the research and interpretations of male and female roles and gender images in prehistory has been occupied with analyses of male and female work and activities, many activities in the past still remain to be problematized from a gender perspective. The reason for this is probably partly that the ambiguous find circumstances make it impossible to establish connections between some activities and either males or females, and partly because there is no specific association between a particular kind of activity and man or woman according to the interpreter's contemporary gender ideology.
It is interesting to note that one activity that is hardly ever mentioned in studies of task differentiation, based on interpretations of grave finds, is the construction of the graves. There are many tasks connected with the burial, the construction of a grave and in some cases the building of markers above the grave of the deceased. The Swedish archaeologist Agneta Lagerlöf (Bennett) has more or less intentionally made way for questions of who built the graves. Bennett has been able to show that constructions of stone over graves in the area around Lake Mälaren, in eastern Sweden, present gender-speciftc traits during the Migration period. Stone constructions over graves with deceased who have been sexdetermined as womenaccording to the artefacts found in the gravesare tidier and Current Swedish Archoeology, VoL 7, l999 have a more distinct design than stone constructions above graves where the individuals have been sex-determined as men. A great part of the women's grave markers consist of stones not more than 0.3 meter in diameter, and Bennett shows that the center of the grave marker varies more in shape among women's graves than among men's graves. A round stone appears to be reserved for women in this geographical area during the Migration period (Bennett 1987:45,13l i).
Bennett does not present any background for the difference between the men's and women's markers of stone that she discusses.  1991:98, 105). Hjorungdal problematizes the traditional meaning of the concept of production. The term "production" has been defined as material production, but Hjorungdai uses the term to describe what she calls the "total production". The total production includes both material production and symbolic production (Hjerungäal 1991:113 overlapping between the sexes that may not be solely due to hereditary biological traits, but also to the skeleton's response to external forces and stimuli such as the performance of tasks (Donlon 1993:98,101). Physical anthropologists of today agree, however, that the best criterion to use for sex determining human skeletons by the osteological method, is based on the shape of the pelvic girdle (Donlon 1993:99;Götherström et al. 1997:73).Differences in the pelvis due to men's incapability to give birth to children and women's ability to do this, are considered to indicate whether the skeleton belongs to a man or a woman. But the pelvic girdle is also an anchor for several muscles of the lower limbs, which means that there is risk that all morphological diAerences in the pelvic region are not a result of differential reproductive function. Some of the characteristics of the pelvis may consequently also be mechanicalfunctional (Götherström et al. 1997:72).The question that follows with these statements is how physically active women or physically passive men are to be recognised in skeletal samples, if there is not an interpretation of a prehistoric gender ideology and/or contemporary gender images that embraces the possibility of a varied and complex task differentiation between men and women.
It appears that the method for reducing presuppositions of male and female activities and hierarchical relations between those-by analysing men and women as individuals, is limited. This is at least the case when the endeavour is to establish an interpretation of a gendered task differentiation. The best way to write about men's and women's tasks would be to let different researchers carry out physical-anthropological sex determinations of skeletal remains and analyse the function or meaning of artefacts among the grave goods separately, and bring the interpretations of sex and task together and formulate a synthesis of them. A synthesis which is sensitive to ambiguities and deviations, and which gives priority to an analysis of multiplicity rather than adjustments of interpreted characteristics for the purpose of making different cases conform to set images and stereotypes.
female in prehistoric task differentiation. Readers may object that the deconstruction and emphasis on variation are typical phenomena of post-modern gender theory, and are just as contaminating as traditional and early feminist research. The fact remains, though, that traditional and feminist research have not been able to break with the stereotyped interpretation oftypical male and female tasks or work. Hopefully the alternative way of viewing differentiation within the gender perspective will increase our knowledge and deepen our understanding of the organisation of tasks during prehistory, as it attempts to avoid simplistic models and generalising concepts.
English revised by Laura 8'rang.

CONCLUSION
This article can be regarded as a contribution to how to begin to think in different ways from previously with respect to task differentiation between men and women. It is meant to show how the gender perspective is able to make way for alternative pictures of male and A CKNO kVLEDGEMENTS Iwould like to thank the Gender-seminar held in Lund on 12" ofApril 1997, Dr. Elisabeth Iregren and Professor Roberta Gilchrist for their comments on the proposals and the ideas in this paper.