Halls of Mirrors: Reflections on the Social Meanings of Early Medieval Rulers’ Residences

The commentators’ reflections on rulers’ residences as a manifestation of developing socio-political complexity raise some issues that require clarification. In our discussion of the temporality of...

The commentators' reflections on rulers' residences as a manifestation of developing socio-political complexity raise some issues that require clarification. In our discussion of the temporality of southern British great hall complexes we point out that the antecedent phases of these sites embody evidence for earlier social and economic complexity, albeit expressed in other, non-monumental, forms. Andrew Reynold's observation that 'lower order settlement seems to come before an elite presence' thus seems to us too simple a generalization, and we feel that Sam Turner perhaps underplays understandings of the fifth-and sixth-century social background to the horizon of monumentality embodied by great hall complexes and princely burials given that detailed models have been proposed for the evolution of social hierarchy and political authority over these centuries. (e.g. Bassett 1989, Scull 1993, 2011, Arnold 1997. Overall, we would argue that the archaeological evidence points towards the late sixth and the early eighth centuries as key thresholds of change in the articulation of rulership. And while John Ljungkvist makes a valid point about the differing time-scales and trajectories over which configurations of power developed in Scandinavia and post-Roman England, we would question whether it is accurate to characterize communities in later sixth-and seventh-century England as 'young' and somehow not rooted in the landscapes in which they lived. These were not rootless immigrant communities (if that is what is implied) and, as discussed further below, conditions of social complexity can exist without a culture of monumentality. Nobody would seriously question Turner's view that aspects of Christian ideology provided social technologies which enabled powerful actors to consolidate and accelerate change, nor that Christian ideology contributed to shaping new landscapes in the early Middle Ages. But the question here is whether, as has been forcefully argued by Blair (2018), the demise of great hall complexes is to be understood primarily as a consequence of the triumph of Christianity and monasticism. We think not: if these were complex multi-modal phenomena then both their inception and demise can only be understood by considering the interaction of multiple influencing factors and by drawing upon wider conceptual understandings of religious transformation that can help to tease apart how contemporary rulers selectively absorbed, reappropriated and -in some casesrejected Christian traditions and influences (Thomas et al. 2017).

REGIONALITY, VISIBILITY AND RECOGNITION
Closely bound up with regionality is the distinctiveness or otherwise of elite architecture within its contemporary cultural context. Speaking of southern Scandinavian hall culture, Ljungkvist notes that architectural scale alone is an unsatisfactory proxy for social gradation which chimes with the observations that we made for early medieval Ireland. This serves as a reminder that the embellishment of halls, external as well as internal, is likely to have played an important role in proclaiming messages of prestige and social difference. Reynolds questions whether some of the southern British sites represent elite residences at all, entertaining the alternative view that they may be storage facilities or simply ordinary settlements. Although challenges attend the characterization of early medieval settlements and their constituent buildings (Hamerow 2012, pp. 98-101, Loveluck 2013, there are serious problems with this argument, not least the inherent contradiction posed by his subsequent observation that great hall complexes largely appear to 'pre-date service features on settlements' and the compelling spatial and architectural similarities between the Thames Valley sites cited and unambiguously residential contexts such as Yeavering. We note that the differences between archaeologically-recognized barns and great halls far outweigh the similarities (Hamerow 2002, p. 37-38, Gardiner 2012, McKerracher 2018. Reynolds observes an interesting distinction in the geographical contextualization of the two papers which invites further reflection. Our examination of the Irish evidence made specific reference to Continental Europe to make the point that previous conceptualizations of rulership in this region have been overly insular and inwardlooking (see also Gleeson 2020). The case is of course very different for Southern Britain where politics, rulership and identity have often been examined within the prism of continental connections and influences (e.g. Wood 1983, Loveluck and Tys 2006, Nicolay 2014. However, it is hard to find visible signs of these connections when one drills down into the specifics of residences themselves. As has often been commented, clear expressions of elite residence before the end of the eighth century are conspicuous by their absence across large swathes of the Continent, and markedly so in areas on the opposite side of the English Channel (Wickham 2005, pp. 500-507, 609, Zadora-Rio 2009, p. 89, Nicolay 2014. Unless the archaeological picture changes dramatically, then it seems very unlikely that these areas provided direct inspiration for great hall complexes, although we acknowledge that continental impulses may lie behind some of the distinctive architectural traits associated with Kentish expressions of the idiom (Thomas 2018). While tempting, it is unsatisfactory to fill the gap by invoking Carolingian Pfalzen as these post-date southern British sites by as much as a century. Such comparisons may have a place in charting persistent tendencies in spatial and monumental articulation of rulers' residences across space and time (Rollason 2016), but are of limited value for discerning historically specific interactions and relationships.

INTERSECTIONS: RESIDENCE, ITINERANCY AND ASSEMBLY
Several of the respondents grapple with the question of how the functions of rulership interrelate spatially and temporally, inviting critical reflection on the de-centralized character of early medieval political geography. Gordon Noble argues that archaeologists have been too ready to accept itinerancy as a universal given of early medieval rulership and we would agree that more attention needs to be paid to historical contingency. In Ireland we have shown that a strong case can be made for viewing the inception of itinerancy and more formalized expressions of elite residences in the eighth century as intimately related. Expressions of a similar interrelationship can arguably be seen in a southern British context in the sixthseventh centuries. For example, the compact geographical cluster of royal vills seen in east Kent, of which Lyminge formed part, could be argued to fit this model, as would the archaeological recognition of contemporary central places in south-east Suffolk.
In discussing the relationship between residence and assembly, Alexandra Sanmark makes the distinction between public/thing assemblies on the one hand and royal councils on the other, the former held in outdoor settings and the latter in the indoor environment of the mead-hall. While this duality pertains widely across Scandinavia, there is evidence for a more negotiated, spatially intimate relationship between these two types of assembly in relation to top-level royal sites. The same seems to be true of Ireland, as testified by the site of Knowth where, as we have seen, the storage of food renders is juxtaposed with ceremonial and residential activity. A similar blurring of lines can also be seen at the southern British site of Yeavering which combines evidence for the infrastructure of public assembly (a timber grandstand) alongside halls designed for smaller, more socially restricted gatherings within its monumental core (Hope-Taylor 1977). While it is difficult to make wider generalizations, it is interesting to speculate whether the agency of great hall complexes lay partly in their embodiment of novel attempts to conjoin public and private space. Additionally, we need to consider the possibility that the wider landscapes of rulers' residences might harbour archaeological evidence for episodic cycles of feasting and other communal activities of the type seen at Scandinavian assembly sites (Semple and Sanmark 2013).
The meanings of rulers' residences and places of assembly may also have converged as mutual foci for the negotiation and materialization of collective identity. Recent studies have placed emphasis on the role played by early medieval assemblies in fostering perceptions of collective belonging and group affiliation (e.g. Semple et al. 2020). Sindbaek (2008) has argued that this process is reflected in spatial correlations between thing-districts and regionalized cultural practices in early medieval southern Scandinavia. Outdoor thing sites have taken centre stage in these discussions, but there is a need to accommodate rulers' residences within such conceptualizations given their attested assembly function. Indeed, as centres of production and innovation, these environments must surely have contributed to the materialization, consolidation and periodic reinforcement of perceptions of collective affiliation.

FUTURE DIRECTIONS: THEORY, DATA AND INTERDISCIPLINARITY
The comments on our papers offer divergent priorities for future study of early medieval rulers' residences. Both Noble and Sundqvist conclude by stressing the importance of further excavation to enlarge and enhance existing archaeological datasets, an aspiration that would undoubtedly pay major dividends for some regions. However, as testified by our review of the Irish evidence, important progress can be made by investing new research in old sites to improve chronological and spatial understanding which, as seen at Lagore, can bring previously unrecognized residential phases to light. Reynolds and Sanmark place rather more emphasis, implicit and explicit, on the need for interdisciplinary landscape studies directed towards clarifying how rulers' residences interacted with outdoor assembly sites and other foci for the performance of power.
We are surprised that none of the responses has engaged more closely with the theoretical agendas presented in the first of the papers (Gabor and Scull). Here, we outlined the potential of a practice-led framework for examining the performative dimensions of rulership and argued that this approach can help to bridge divisions in previous scholarship by providing a common prism for investigating the regimes and 'ways of doing' woven into the political, mythic and ideological dimensions of these sites. At the end of our paper we highlighted how a practicebased framework can inject new interdisciplinary vitality into the interpretation of rulers' residences, particularly with reference to reconstructing the embodied experience of elite space and to understanding the relationship between architecture and innovation. Reynolds shares our view that the experiential dimension of these sites is worthy of future attention, supported by highresolution reconstructions to enhance archaeological interpretation. However, we envisage an altogether broader and more collaborative model: engaging experts in architecture, construction management and related studies to direct the sophisticated, human-centred methodologies used in these fields towards enriched interrogations of the early medieval built environment. By providing a conceptual language that speaks across disparate disciplines, a practice-based approach holds the key to such future endeavour.