Early Medieval Memories

Early Medieval Memories: the Past in the Past

The study of the ‘past in the past’ – the perception of earlier historical, legendary and mythical times in the past – has been the focus of archaeological investigations into how social memory operated in past cultures. For the early Middle Ages in Britain, this has been a rich focus of research over the last two decades. This research theme provides the background to the importance of the fieldwork investigating the Pillar of Eliseg.

The Past in the Early Medieval Past

Early medieval burial sites and funerary monuments often deliberately reused earlier prehistoric and Roman-period sites and monuments. Richard Bradley (1987) argued that rather than real continuity, we are seeing a deliberate strategy of inventing traditions at such sites, in which the dead and the living were situating themselves in connection to a timeless, mythical past. Ancient monuments may have been regarded as the residences or burial places of heroes, ancestors, ancient peoples or supernatural beings. In this sense, ritual practices were used in the early medieval period to create senses of place and history in the landscape through monument re-use.

A number of studies have attempted to build upon Bradley’s insights, emphasising the selectivity of monument reuse, both in terms of which monuments were adopted and which were avoided, but also in terms of the exclusive selection of particular monuments for particular categories of the dead or particular monument-types. Across early medieval Britain and Ireland, early medieval burial sites and monuments are found appropriating earlier monuments in the landscape.

Why Monument Reuse?

It has been argued that the re-use of an ancient monument evoked particular conceptions of time and memory and providing a communal focus for a cemetery (Bradley 1987). Such links to perceived ancestors and supernatural beings, or indeed, perhaps the deliberate appropriation and invention of new pasts through the appropriation and ‘forgetting’ of earlier traditions, may have had a range of social and political motivations. The desire to legitimise claims over land and resources and thereby sustain inheritance claims and territory may have been a chief motivation (e.g. Lucy 2000; Shepherd 1979). Yet this author has argued that the practice concerns the working and re-working of social memory at a variety of levels: links to ancestors and the sacred in a general sense but also the creation of specific genealogies and histories for kingdoms and communities (Holtorf 1996; 1997; Williams 1997; 1998; 1999b). This was not simply a concern for the past, but also a means of staking claims over the present and the future (Holtorf 1996; Penn 2000); using the past as a meanings of appropriating and shaping the early medieval landscape both conceptually and physically through mortuary practices.

While the desire to place the dead close to pre-existing monuments can be demonstrated to be a repeated and influential factor in the location of early medieval burial sites prior to the adoption of churchyard burial from the fifth century through to the eighth and ninth centuries, attitudes towards ancient monuments were neither static nor universal. John Blair (1994) has argued that the seventh-century was a particularly important time for monument reuse because of the tensions and transformations in society, economy and religion at this time. Moreover, Sarah Semple (1998) has argued that after the seventh-century, the ancestral or communal associations of old earthworks became augmented and supplanted with less positive associations. Archaeological evidence demonstrates an increase use of mounds as places of execution and criminal burial from the seventh century onwards. Place name, literary and pictoral evidence support the view that in later Anglo-Saxon England, selected earthworks (including both ‘pagan-period’ burial mounds and prehistoric monuments) could receive demonic and supernatural associations (Semple 1998).

Within this broad picture, there is a need for more contextual analyses of monument reuse, focusing on localities and regions. This approach has been developed by Sarah Semple who has looked at monument re-use in the Avebury region. She recognises the preference for single, intrusive graves selectively inserted into large prehistoric round barrows although the communal re-use of monuments is also known from the region (Semple 2003). This practice began in the sixth century but develops into the seventh and eighth-centuries and possibly into the ninth century. Both genders are found buried in older monuments, but contrary to broader trends, there is a bias towards male interments in the Avebury area (Semple 2003: 74). Over time, Semple sees topographical changes, with the West Overton burials of sixth-century date indicating re-use close to contemporary settlements in a densely populated dowland landscape. Meanwhile in the seventh century, the secondary burials are placed in prominent locations on the chalk escarpment with extensive views further away from the likely locations of contemporary settlements at the base of the escarpment (Semple 2003: 76; see also Lucy 1998).  Semple relates this pattern in relation to the increasing political contestation of the region in the seventh and eighth centuries between the kingdoms of Mercia and Wessex, a possible context that might also explain the building of the Wansdyke; a linear earthwork traditionally dated to the fifth century AD (Reynolds 1999). Similar localised stresses might explain the long-term but varied adoption of ancient monuments in early medieval Wales.

Monument Reuse in Early Medieval Wales

Early medieval monument reuse was not primarily an ‘Anglo-Saxon’ phenomenon however, it is widely recognised as a commemorative strategy for early medieval cemeteries and Class 1 inscribed stones across early medieval western Britain, including Wales.

Capel Eithin, Anglesey

The early medieval cemetery at Capel Eithin on Anglesey is one of the most extensively excavated ‘early Christian’ cemeteries from western Britain. Excavations by Gwynedd Archaeological Trust revealed a cemetery of west-east graves, many in cists. Human remains were poorly preserved, but in total 102 features were identified that were of a size, shape, orientation and internal composition to suggest that they were graves (White & Smith 1999: 128). The graves focused upon one rectangular structure that not only seems to have been an important monument marking a ‘special grave’, but it also operated as a focus for subsequent burials. The structure consisted of a rectangular trench with straight sides and a flat bottom. On the northern side were traces of organic remains within the trench that may have been the remains of a wooden beam while a scatter of stones of a type that had to be imported from 8km to the north were found scattered within the southern trench. One possible interpretation is that a wooden structure of logs, planks or even wattle was supported by the trench with the stone used as packing for the footings (White & Smith 1999: 136). The structure had a narrow eastern entrance the size of a doorway, 0.80m across and thin layer of clay provided a floor surface for the structure. It was thought to have been a roofed structure rising from the clay floor and accessed from the eastern doorway.

The structure enclosed three graves. Grave 66 was placed centrally within the enclosure, lined with stones in a comparable manner to many other graves from the cemetery. Of the two instances with substantial traces of organic remains, one was grave 66 (White & Smith 1999: 130). The stain was shorter than the grave suggesting a grave structure or plank rather than a coffin covered the body.

Two further graves were associated with the Capel Eithin structure. Grave 67 was placed within the enclosure to the south of grave 66 and from its small size it may have contained the body of an interred child. Both graves 66 and 67 were subsequently covered by the clay floor surface, suggesting that they were integral to the earlier phases of the enclosure’s use. At a later date, an additional cist grave was added to the doorway area of the structure, cutting the top of grave 66 but seemingly respecting its location and that of the still-standing ‘mortuary house’. Perhaps the aim was to ‘seal’ the doorway with the new grave and prevent further access to graves 66 and 67. In any case, the structure served as a focus for one of the three main burial clusters in the cemetery. These graves had the largest grave dimensions and contained the most frequent use of complete cist graves from the site, hinting at the possibility that this was a preferential burial site for a higher status group (White & Smith 1999: 140).

The interpretation of such structures as a cella memoria – a memorial cell raised over the grave of a ‘saint’ or secular leader – was seen by the excavator to be likely, with the proviso that the lack of artefactual evidence suggesting that it was not used for any length of time (White & Smith 1999: 158). However, if the structure was repeatedly cleaned and its structure repaired, the ephemeral ground-level remains may still represent an enduring structure used for years or even decades. Following the lead of Nancy Edwards (2002) we should be wary of seeing these as necessarily the graves of ‘saints’ or members of the religious hierarchy in early medieval Wales because they may equally have marked those attributed to the graves of those with a special secular status. Moreover, just because in later centuries these graves may have become the focus of special attention, this need not reflect the accurate transmission of memories concerning their original occupants. Whoever their occupants were, the structure fits with other mausolea and cellae memoriae known from the late antique Mediterranean, western Europe and other instances identified in early medieval western and northern Britain (Petts 2004). The structure facilitated commemoration by provided continual access to the grave and subsequently attracting further graves. Although we cannot be sure whether this was the first burial at the site, or one constructed later in the cemetery’s history of use, it is evident that it formed part of a cluster of graves, suggesting that the location had a special importance for the community long after the initial burial.

Further evidence for the significance of the site comes from an early Christian inscribed stone. Edwards accepts that many of Nash-Williams’ Class I inscribed stones were originally funerary monuments, if not raised over graves they were certainly, in part, commemorating (mainly male) ancestors (Edwards 2002). Capel Eithin was associated with such an inscribed stone, albeit now lost, but recorded in a manuscript of c.1698. The excavators even speculate that the stone may have come from the area of the excavated cemetery, perhaps set into one of the stone-packed pits identified during excavation. The Latin inscription reads DEVORIGI, a Celtic name that Patrick Sims-Williams ascribes to the sixth or possibly the seventh century AD. An inscribed memorial stone was also found associated with the early medieval cemetery at Arfryn, Bodedern (also on Anglesey), forming the lintel of a grave. By analogy, the cemetery may have focused on a number of graves commemorating high-status individuals whose remembrance, enhanced by both inscribed stones and memorial cells, encouraged the continued use of the site for burial over a number of centuries.

Plas Gogerddan, Ceredigion

A second Welsh site provides evidence of related rectangular structures surrounding early medieval graves. At Plas Gogerddan in the Afon Clarach valley north-east of Aberystwyth, excavations focused on a Bronze Age ritual complex including round barrows and standing stones that subsequently received burials in the late Iron Age. The excavations also revealed evidence for an early medieval cemetery of west-east alignment and arranged in rows. Nine graves produced surviving coffin soil-stains but little human bone survived in the soil conditions. The site was broadly dated to between the third and seventh centuries AD on the basis of a single radiocarbon date (Murphy 1992: 15-17).

At Plas Gogderddan, three rectangular structures were identified, the best preserved of which (structure 373) consisted of a rectangular foundation trench with an opening at the eastern end bounded by two post-holes. Within the enclosure was a central grave, but also a stone-lined pit. The dark soil stain, hinting at the presence of decayed timbers, ran throughout the trench. As at Capel Eithin, we may be seeing here evidence for a timber structure supported by the trench. At the eastern end was a gap, 1.8 m wide, with two small post-holes set 1m apart, creating a narrower aperture for entering the structure. The excavators argue that the stone-lined pit, being aligned on the grave and surrounding structure, was an integral part of the funerary ritual. One suggestion broached by the excavations is that a wooden box or comparable organic container had been placed in the pit, perhaps as grave goods or offerings; certainly phosphate analysis confirmed that no body was ever inhumed in them. As with Capel Eithin, it appears that the unenclosed burials cluster around the structures, hinting that they formed commemorative foci for both funerals and other ritual even

Tandderwen, Denbighshire

A different form of monument was identified during excavations at Tandderwen in the Ystrad valley of north Wales (Brassil et. al. 1991). Aerial photographs revealed a series of square ditched structures that upon excavation were shown to enclose graves of early medieval date focusing upon a large ring-ditch, the surviving traces of a Bronze Age barrow. The site incorporated a mixture of monument reuse and monument-building as an integral part of the funerary rituals. The Bronze Age round barrow was enclosed by an enigmatic rectangular structure 576 on the southern edge of the excavated area. The structure consisted of a four-entranced square enclosure encircling the prehistoric burial mound. The entrances were positioned half way around each side, marked by rounded butt-ends to the ditches. This feature was excavated after a prolonged silting of the Bronze Age ditches and an early medieval date in association is not unreasonable but cannot be proven (Brassil et. al. 1991: 61-62). Whether this was a trench for an above-ground fence or earthen bank, or simply a way of demarcating and respecting the earlier monument is not clear. Yet it does represent a renewed interest in the ancient monument as a focus for ceremony and ritual connected to the graves. It may even represent the deliberate choice to appropriate an old structure that may have been invested with supernatural or ancestral associations in a fashion that would not have been regarded as appropriate for a new cairn or mound. It is even possible that the barrow itself were remodelled and re-used by early medieval burials that have left no trace as the burial mound was eroded by subsequent centuries of weather and farming activities. At the very least the excavation plan clearly suggests that the ditch or mound of the original barrow were still visible in the early medieval period and the cluster of graves to the north suggests that it provided a commemorative focus for the cemetery (see chapter 7).

The cemetery itself lay north of this monument, consisting of 39 west-east oriented graves, some placed in rows and groups of between two and six graves. Human remains had been completely destroyed by the soil conditions. Traces of wood from 22 graves suggest the former presence of coffins of some kind. Individually or collectively, many of these graves could have originally had above-ground monuments raised over them but no evidence survived. However, interspersed between these groups were six graves surrounded by square ditches. In two further instances, (28 & 574), the line of the enclosure ditches were broken half-way along their eastern side comparable to the ‘special graves’ identified at Plas Gogerddan and Capel Eithin discussed above. In one of these cases (28), the enclosure was of a substantial size, demarcating a sizeable space 9m by 8.5 m around the grave (Brassil et. al. 1991: 64). Rather than representing the footings of small buildings, it is more likely that these ditches marked the inside of an outer bank, or perhaps most likely, an internal mound. The excavator notes that the thickest ditch was associated with the only structure containing three graves (574). One scenario is that the mound was reconstituted after the second and third interments by re-cutting the ditch and throwing up a new mound on each occasion. The evidence hints at a succession of mortuary events, each connected to mound-building episodes, leading to augmentation of the monument (Brassil et. al. 1991: 64).

The Pillar of Eliseg in Context

In conclusion, archaeological evidence is revealing the reuse of ancient monuments as a widespread and key commemorative strategy by which early medieval communities defined and negotiated their relationship with the past to constitute and bolster their identities and claims over land and other key resources. Project Eliseg seeks to enhance our understanding of the ideology of monument reuse in early ninth-century Powys, as one manifestation of a broader phenomenon.


NOTE: The above text is adapted from sections of Howard Williams’ 2006 book Death and Memory in Early Medieval Britain published by Cambridge University Press.