6.2 The significance of the sunken-featured buildings

The results and interpretation of the integrated geophysical survey add a number of new points to our knowledge of the archaeological remains at Aggersborg. In particular, the observed distribution of probable SFBs indicates that the total number of such buildings in the site is more substantial than could be seen from the excavations. In the excavated areas most of the c. 150 SFBs were clustered in a single area, in which remains of post-built houses and other features relating to the pre-fortress settlement were also concentrated. However, the evidence from the surveys, combined with a re-analysis of the early excavations, strongly suggests that, in addition to the main concentration noted in the excavations, there were further clusters of SFBs, reaching a total number of as many as 350 buildings. The many newly detected SFBs also raise the possibility of further, undiscovered, structures, including post-built houses, which were not detected clearly in the results.

A further important observation concerns the chronological development of the site as indicated by the newly defined distribution of SFBs. The conclusion that a number of buildings in the previously little-known cluster to the west of the fortress may relate to activity that occurred after the fortress was abandoned indicates that the practices that led to their construction were more long-lived than had been previously suspected, and survived the building as well as the abandonment of the ring fortress.

The number of SFBs suggested by the surveys sets Aggersborg apart as a rare site. In many contemporary settlements the number of SFBs is much lower. From the nearly completely excavated villages of Vorbasse and Sædding, where the number of post-built buildings excavated is many times higher than at Aggersborg, the total number of SFBs was well under one hundred (Stoumann 1980, 111; Hvass 1980, 147; 1988). More substantial numbers of SFBs are sometimes noted at coastal sites (Henriksen 1997; Ulriksen 1998; 2006; Svanberg and Søderberg 2000; Callmer 2002). While excavated examples rarely exceed a few dozen in any one site, aerial survey and geophysical prospection have in some cases indicated over a hundred buildings on one site (Eriksen et al. 2009, 60; Dobat 2010, 176).

Two sites in the Limfjord region have produced similar numbers to Aggersborg. At Bejsebakken, near Aalborg, some 350 SFBs were excavated in extensive campaigns during the 1990s, presumably comprising almost all buildings on the site; some 80 buildings excavated at Sebbersund near Nibe, about mid-way between Aalborg and Aggersborg, are suggested by aerial survey to form part of a cluster of some 300 buildings in total (Nielsen 2002a, 11; 2002b, 200). Like Aggersborg, both these sites are central locations in the coastal landscape: Sebbersund is situated in an excellent natural harbour, while Bejsebakken is at an old crossing point. The nature of these locations supports the view that traffic potential was a factor that contributed to making these places suitable for the activities to which the SFBs pertained.

Various models may be considered to account for this observation. Substantial clusters of SFBs are sometimes encountered in coastal sites with few or no traces of post-built buildings. Some may be specifically related to maritime activities including the building and maintenance of boats, and almost certainly to direct exploitation of the resources of the coastal landscape through fishing, gathering and animal grazing in coastal meadows and forests (Ulriksen 1998, 159f). Sites of a similar, albeit smaller, appearance in inland locations are suggested to be shieling sites used by herders for summer grazing (Sindbæk 2012). Among object and fauna assemblages from Aggersborg, many finds bear witness to fishing or other use of marine resources: fish-hooks, eel-forks, fish-bones and shells were noted in many contexts (Roesdahl 2014; Hatting and Rosenlund 2014). No evidence of more extensive sea-going activities, for example the building and repair of boats, was found here. Moreover, with the possible exception of an early (9th century) phase of the settlement, during which house D is the only large building identified within the excavated areas, the SFBs at Aggersborg co-exist with traces of a broad-based rural settlement.

In many rural settlements, SFBs are, with great probability, related to activities practiced in the individual farms. This may also be the case for some buildings in Aggersborg, and the clustering of buildings in the same area as post-built houses from the pre-fortress settlement is consistent with such a model. Each of the several farms indicated by the main post-built houses could be expected to have included a few SFBs, leading to an accumulation of remains as buildings were abandoned and renewed over the course of a couple of centuries. A high number of SFBs in coastal settlements could be surmised to relate to activities that were more intensively pursued in coastal sites, through access to additional resources or potential for exchange. Even by the most generous count, however, this model leaves a substantial number unaccounted for.

It has recently been suggested that some sites with many SFBs represent accommodation at regional assembly sites. Jørgensen et al. (2011) draw parallels to the north Scandinavian church-villages (kirkebyer) of the early modern period, in which a widely dispersed rural population of sub-arctic regions maintained cabins for gatherings at important church festivals. They observe that groups of SFBs associated with elite sites like Tissø, Sjælland, could have acted 'as temporary accommodation for families who attended events at the site for short periods.' (Jørgensen et al. 2011, 103).

While it is highly likely that Aggersborg attracted periodic gatherings in the Viking age, visitors to regional assemblies in Denmark in later periods would usually manage their journeys as day-excursions. If sites with many SFBs like Aggersborg, Sebbersund and Bejsebakken were all assembly sites, no one in their hinterlands had a need to ride or sail for more than 20km, and most would have had a considerably shorter journey. In this respect the analogy with the north Scandinavian kirkebyer, which served populations spread over hundreds of kilometres, is weakened by an obvious difference in scale.

A further complication is the context of the buildings at Aggersborg. Key to the model of SFBs as assembly accommodation in Tissø are the observations that evidence of activities mainly point to mobile tasks (e.g. whorls for spinning), while 'little or no evidence of ordinary agrarian structures and functions' is noted in the site (Jørgensen et al. 2011, 104). Aggersborg is different: not only do the SFBs exist within a busy rural settlement, but the finds from the bottom layer of a number of SFBs include loom-weights from upright warp-weighted looms, hardly a piece of equipment that would have been brought and fitted for a few days' excursion.

In order to explain the high number of SFBs at Aggersborg it would seem more promising to turn to another suggestion proposed by Jørgensen et al., according to which SFBs acted as temporary accommodation for purposes other than assemblies, in particular for 'the time-consuming processes of turning plant fibres into cloth', which required 'housing for the families or staff … from other settlements' (Jørgensen et al. 2011, 105). It may be suggested, then, that the SFBs at Aggersborg are not related directly to assemblies, but that both stem from a common factor: the existence of a high-status farm.

The character of SFBs points to use as workshops and shelters, but with no traces of heating or basic domestic activities like food preparation, they must thus be related to activities that called for accommodation and work space, but not a permanent settlement. In the context of an elite residence, as indicated by Aggersborg's large house D, it can be suggested that such activities were pursued by people who were obliged to render service to their lord, perhaps for a specified number of days in the week, or for particular periods in the year. In an early medieval Scandinavian society where land rent was probably not yet firmly established as an institution, such corvée work was among the basic rights that lords might possess over their subordinates (Iversen 2005; Helle 2006, 92; Poulsen 2012, 278). The evidence for textile working associated with SFBs at Aggersborg is paralleled at a number of other sites in South Scandinavia (Thomsen 2009, 506; Sørensen 2011, 50), and ties in with similar evidence across northern Europe (Hamerow 2002, 32f). Textile production would have been an essential way of converting lordship over resources and labour into a profitable product.

As a pattern of settlement at Aggersborg this model accommodates several key characteristics of the site: the combination of many SFBs with several farm buildings and one particularly large house; the situation at a central location, with access to a rich resource area, able to feed a large population; and the continued construction and use of the SFBs after the building and abandonment of the fortresses, which may be interpreted as evidence of a continuation of the work obligation, now – if not previously – certainly tied to the crown estate.