5.2 What caused the shift to a triadic strategy during the Late Iron Age?
5.2.1 Regions and microregions in Late Iron Age Scania
During the onset of the Late Iron Age (Migration-Vendel periods), the mixed triadic system was introduced. It was not fully established, but existed in rudimentary form, visible in more mixed and varied zooarchaeological signals from the studied settlement.
Spatial patterns were hard to detect, but the following observations were made: a tendency of higher abundance of pig/cattle bones in the Kristianstad area; the north-east/south-east manor sites (Järrestad and Hammar) characterized by higher-than-expected abundances of pig bones; coherency surrounding the Uppåkra region appears, with signs of specialized settlements in its vicinity in terms of animal production. This is mirrored in other archaeological material, e.g. the ‘horse site’ Knästorp was also a location of fine metal production in foremost bronze (Stark, 2018: 17). Another example in this area is the ‘sheep assemblage’ Lilla Uppåkra, which has been interpreted a more “ordinary” part of the Uppåkra complex (Magnell, Boethius and Thilderquist, 2013: 90; Lenntorp, 2013).
Helgesson (2002) highlighted Uppåkra as a central region, and it seems that it was coherent in terms of animal production organization as well. In fact, several scholars argue that Uppåkra was the main power centre of Scania (e.g. Helgesson, 2002;
Söderberg, 2018; Larsson et al., 2020). The zooarchaeological and archaeobotanical materials both point to strategies related to centralized settlements (Magnell, Boethius and Thilderquist, 2013; Larsson, 2015), such as the import of cattle (Price, 2013).
Strontium isotope signals of cattle, pig and horse, as well as barley, wheat, oat and rye at Uppåkra and the nearby settlements of Uppåkra 37, Stanstorp and Uppåkra 2:25 indicate a complex strategy of moving both livestock and grains between regions (Larsson et al., 2020). While Larsson et al. (2020) show that pigs probably were bred locally at Uppåkra, they confirm earlier results that cattle were to some degree imported to Uppåkra (see Price, 2013). Larsson et al. (2020) focus their discussion on the centrality of Uppåkra, and an apparent network that included grains and livestock. This could have many explanations, such as the contribution of cattle in large-scale consumption, or agricultural trade (see also Magnell, Boethius and Thilderquist, 2013).
The nearby settlements Uppåkra 37 and Stanstorp also have indications of non-local input of cattle (Larsson et al., 2020). The archaeology indicates the presence of comparatively wealthy farms at these sites (Söderberg, 2018; Bolander and Söderberg, 2019). What does the non-locality mean for these sites? Does it mean that the richness of Uppåkra ‘spilled’ onto the nearby settlements (Larsson et al., 2020) or does it relate to other factors? The recent study on Uppåkra area by Aspeborg (2021) provides a long-term perspective, important for better understanding this issue. The establishment of the central area, with Uppåkra as central place, in the Roman Iron Age, was clearly related to the changes, visible in settlement patterns, in the late pre-Roman Iron Age
(Aspeborg, 2021). The establishment of Uppåkra, and the period preceding this, was probably politically very dynamic, which led to an organization based in trade and exchange with other regions and within the region. This would also explain the clear specialisation in the different settlements of the area, as well as the gathered wealth exemplified not only at Uppåkra but also at Stanstorp, Knästorp and Uppåkra 37.
According to the above review, Uppåkra and its closest surroundings were, to some extent, dependent on import of mainly cattle (and grains) from the surrounding land (see Magnell, Boethius and Thilderquist, 2013: 121). Strontium isotope analyses show that most cattle were locally bred, but also that some were not (Larsson et al., 2020).
Input of animal import is indicative of higher degrees of centralisation (Macheridis, 2017c). The research clearly show intense agricultural trade and production in this region. This is not seen in other parts of Scania. On the other hand, the Uppåkra region is a focus for much detailed research, for example, when it comes to strontium isotope analyses. The main difference in animal production seem to be between south-west and east Scania. It supports the view proposed earlier that there was cultural and sociopolitical differences between these areas (e.g. Fabech, 1993; Helgesson, 2002). It might also mirror different aspects of centrality and site functions. The Järrestad complex was characterized by open pasture. Järrestad itself was a manor site, with political and ideological functions (Söderberg, 2005). The possibility of rearing cattle, and horse conceivably for ritual purposes (see Nilsson, 2003), is a sign of the prosperity of the site. Perhaps, the foundation of Järrestad’s richness laid on large mammals and provision of meat for ritual and social consumption events (see Söderberg, 2005: 272-273). Strontium isotope analyses have benen made on pig (two samples) from Järrestad, the results indicating that they were locally bred (Ahlström Arcini, 2018: 96). Further, more detailed analysis grains or animal bones would test the hypothesis of the organization of agricultural trade by Larsson et al. (2020). Gårdlösa is another settlement with a focus on cattle husbandry (Gejvall, 1981). How did Gårdlösa and Järrestad agricultural trade networks relate to the ones observed at e.g. Uppåkra? Can we link different productions in terms of export/import to different microregions? This remains to be investigated. There are other indications to take into consideration in relation to this topic, such as specialized production. For example, excavations at Uppåkra produced evidence of beer production (Larsson, Svensson and Apel, 2019); at Järrestad, the finds of hops were interpreted as tied to possible local beer/mead production (Lagerås, 2003: 256).
Turning to the Vendel-Viking settlements, the picture changes. More settlements appeared along the south coast. A higher abundance of horse bones are seen at sites, with the presence of upper social strata. The most notable change is the appearance of site clusters with mixed zooarchaeological signals. Zooarchaeologically, we see the appearance of e.g. the Mossby complex, possibly including Ystad, perhaps an area
around Skegrie, a site with high sheep and marine input, the south-east area with Gårdlösa (cattle), Valleberga (horse), and Järrestad (pigs and cattle). The previously established Uppåkra region is still relevant, with high variation between the sites indicating different specialisations in animal production. Around Löddeköpinge, a more independent region formed, together with Dagstorp and Västra Karaby.22 These areas correspond roughly to the hot spots visible in Strömberg’s analysis, with the exception of the inner midland, the uppermost north-east and the north-west of Scania (Fig. 3). The Löddeköpinge area relates probably to the hotspot surrounding Landskrona in Strömberg’s isarithm map. Löddeköpinge was not discovered in its entirety in 1961.
The zooarchaeological material also gives a perspective on the earlier divisions of Iron Age Scania. The argument that south-west Scania (and Denmark) differed regionally and culturally between the south-east and the north-east remains relevant from a zooarchaeological point-of-view (see Fabech, 1992; Helgesson, 2002; Svanberg, 2003b). The zooarchaeological picture shows, however, that the complexity within each region, or perhaps even within the larger interregional coherency of Scania/eastern Denmark, was high. Small and highly organized hubs of economic independent strategies, in terms of animal production, appeared. I call these hubs microregions, as a term signaling some level of coherency, production organisation and socioeconomic independence.23 In this aspect, I follow Svanberg’s reasoning that “[c]ommunities sharing collective cultural practices do not necessarily correspond to political communities” (Svanberg, 2003a: 149). Some microregions are clearly important on a more interregional level, such as Uppåkra. Some microregions may not have contained sites with similar central importance, e.g. Ystad/Mossby, indicating different local strategies. In this area, however, there are signs of trade contacts in the form of Slavic pottery (Olausson and Larsson, 1985; Bolander, 2017). Based on this, Bolander (2017:
52-53) argues that a market place, similar to e.g. Löddeköpinge, was located within the Mossby complex. The different site functions within each microregion is not elaborated here, except for the above note on e.g. interregional centrality. This requires a more holistic view, incorporating other archaeological categories, and is therefore better suited to future studies. Further, it is important to consider the catalysts for the emergence of these smaller regions. I previously highlighted agrarian transformation in conjunction with social dynamics in the respect. Perhaps these clusters reflect the
22 The ringfort at Borgeby, in the vicinity of Löddeköpinge, was also part of this region. It has been dated to the 10th century CE; during the 11th century, it was under royal rule (Svanberg and Söderberg, 2000: 18, 339).
23 Microregions is as an arbitrary term. Though it has been useful for this study, I do not offer any quantitative criteria for the term. Instead, it is used as an heuristic tool for the discussion.
establishment of highly organized microregions that had an inherent division and organization of animal production, clearly adapted to both environmental factors, such as coastal meadows suitable for grazing, as well as societal structure, such as horse-keeping and consumption in sites with the presence of an upper social strata.
On the Viking Age-Early Medieval transitional sites, pig bones are most abundant.
This is discussed above as related to centralization processes, amongst other things, but it could also be tied to the increased/intensified crop farming (see Lagerås and Magnell, 2020). The clusters seen for this period partly overlaps with the ones from the Vendel-Viking periods, as above. Some relevant patterns in terms of animals are noted: most pig sites are located in north-east Scania and in the Malmö region. Most sheep sites are found along the south coast and south of Malmö, such as at Hötofta/Östra Grevie.
Some cattle sites cluster north of Lund, e.g. Örja. Still, most sites show clear “mixed”
signals, considering the three most common domesticates. The triadic shift was established during this period, as was the crop rotation system, probably in tandem, coinciding with higher centralization, eventually turning to urbanization and state formation processes. Many market places established during the Late Iron Age continued into the Medieval era (e.g. Cinthio, 1975; Callmer, 1986), and so did some of the microregions detected in this study, e.g. Ystad and Lund (former Uppåkra).
Many of the transitional Viking-Medieval sites included in this study formed into the historical villages (Swe. byar) during the Early Medieval period (Schmidt Sabo and Söderberg, 2019: 43). This discussion ends here, as delving into the transformation of the political structure with e.g. state formation is beyond the aims of this study (see e.g.
Näsman, 1998; Söderberg, 2005: 403-441). The Early Medieval villages were also based on mixed farming strategies. Animal husbandry and crop cultivation interplayed in a system with infield farming and outland pasture areas (e.g. Emanuelsson et al., 2002: 54-55; Lagerås and Magnell, 2020). Even if systemic socio-political change did occur, the mixed stock-keeping strategy was successful and resilient from the Late Iron Age throughout the Medieval and post-Medieval periods.