A Research Framework for the Archaeology of Wales

Northwest Wales – Early Medieval






The period c AD 400-1075 in NW Wales, spanning the centuries between the end of Roman rule and the first Norman intervention, is still poorly understood. The documentary evidence is notoriously sparse and frequently difficult to interpret: yet it continues to provide a basic chronological framework. It suggests the possibility of Irish settlement in the NW and the movement of Cunedda and his followers from Manaw Gododdin around the end of the Roman period. It indicates the establishment of the important kingdom of Gwynedd by the early 6th century and its gradual rise at the expense of probable minor kingdoms such as Rhos and Meirionydd. It is also possible to chart to some extent Anglo-Saxon alliances and hostilities and from the 9th century onwards Viking raids and likely settlement. At the same time the documentary evidence can throw light on the development of Christianity by helping to identify major ecclesiastical sites. In the past the documentary evidence has tended to set many of the archaeological agendas. For the future it is the purpose of this paper to set an archaeological research agenda. However it is essential that the crucial role of cross-disciplinary research for this period is also recognised.



A.  Settlement


In 1934 the hut group at Pant-y-Saer (Ang.) was heralded as the first possible early medieval settlement in Wales because of the excavation of a penannular brooch. This was followed in the late 1950s and 1960s by minor excavation of the hillforts at Dinas Emrys and Degannwy, the early medieval phases of which were recognised by the presence of imported pottery. Over the last 25 years or so evidence of early medieval activity has come to light by chance on a few hut-group sites such as Graeanog and Cefn Cwmwd. The most spectacular discovery has been the Viking Age and earlier settlement at Y Glŷn, Llanbedrgoch. The recent research on Llys and Maerdref sites may also be of significance for the early medieval period.



1.  Strengths in our understanding of the settlement of this period are extremely difficult to identify since so few early medieval sites have been recognised.


2.  The discovery and partial excavation of the productive site at Y Glŷn, Llanbedrgoch might be considered as a strength since it is the only site of this kind to have been recognised in Wales and one of the very few rural Viking settlement sites to have been recognised outside Scotland. The fact that there is earlier high status occupation on the same site is also significant.



1.  Although hillforts are clearly a characteristic high status settlement type in the 5th-7th centuries and possibly later as evidenced by the discoveries at Dinas Emrys and Degannwy, it is still impossible to recognise other hill-fort sites in NW Wales of this date and to differentiate them from defended hilltop sites of other dates.


2.  The fact that the productive site at Y Glŷn, which is located on good arable land and was initially located by metal detecting, is unique is a problem. It is likely that other such sites exist which have not been recognised perhaps partly due to the paucity of ploughed land in NW Wales.


3.  Although the excavation of hut group sites has repeatedly hinted at early medieval activity on some of them through the discovery of occasional artefacts and scientific dates, the nature, extent and chronology of such activity has so far proved very difficult to define. The recognition of clearly early medieval structures rather than the continued use or reuse and modification of Roman period ones also remains a problem.


4.  Our inability to recognise characteristic timber structures is a serious weakness since it is likely that wooden buildings, both timber halls as suggested by Y Glŷn and wattle huts as hinted at by Gerald of Wales, would have been comparatively common.



5.  The lack of diagnostic early medieval artefacts, especially native ceramics, makes identification of sites more difficult, especially those of low status.



1. The main aim over the next decade should be arrive at a point where we are able to recognise a range of characteristic early medieval settlement types of both high and low status, have some idea of their chronology and changes in settlement types over time and some awareness of potential regional differences.


2.  The significance of early medieval hillfort settlements in NW Wales needs to be re-evaluated. First, we need a better understanding of sites of definite early medieval date, their structures, artefacts, potential environmental and dating evidence. There has been no major investigation of an early medieval hillfort in NW Wales to match, for example, Dinas Powys (Glam) or Dunadd (Argyll). A major research project on the probable multi-period site at Degannwy incorporating aerial photography, geophysical survey including the ‘saddle’, targeted excavation and the setting of the site within the context of the surrounding landscape and its resources would go some way to answering this. Second, we need to target other hillfort sites to see whether early medieval examples can be identified by their typology or through other means. This should be achieved through aerial and geophysical reconnaissance followed up by strategic excavation to obtain dating evidence.


3.  In order to gain more understanding of high-status settlement at the end of the period more investigation of the origins of llys and maerdref sites is essential. Further targeted excavation at Rhosyr, for example, in order to uncover any earlier timber (?) phases and determine their dating would be of great value.


4.  Further targeted investigation of hut groups would have the potential to enable the discovery of well-preserved examples with early medieval dating evidence. The widely held view that there is a change from round to rectangular structures during the early middle ages also needs to be tested.


5.  In western and northern Scotland and northern and western Ireland the potential of coastal middens and sand-dune sites has been recognised as evidence for early medieval settlement. Although sand-dune sites have potentially been recognised in SW Wales, no settlement sites of this type have been identified in NW Wales. If such sites could be identified and sampled for dating evidence, they offer considerable opportunities for identifying settlements and associated activities.


6.  Ongoing place-name research may have the potential to help identify early medieval sites.



1.  Most of the early medieval population would have lived on the more fertile lowlands of NW Wales, notably Anglesey, the Llŷn and Dyffryn Conwy. These are also the areas of the most intensive modern settlement and farming with threats such as deep ploughing and the potential destruction of fragile archaeological evidence.


2.  Y Glŷn was discovered through the responsible reporting of metal detector finds. No other productive sites of this kind have been reported by metal detectorists but this does not mean that they do not exist. The maintenance of the metal detector finds reporting programme is therefore essential.


B.  The economy, land use and the exploitation of natural resources


It is generally agreed that there was a mixed farming economy in the NW as elsewhere in Wales during the early middle ages and that the best use of pasture was probably facilitated by transhumance. However, in contrast with S Wales, there is no evidence concerning landed estates in the documentary record. The farming economy must have been sufficiently developed to have produced some excess in order to support the life-styles evidenced on the few high-status sites that have been recognised. In the 5th to 7th centuries there is evidence of long distance trade in the form of imported pottery and potentially glass. In the Viking period there are silver hoards with exotic items but the economy seems to have been insufficiently developed to promote the establishment of either emporia or towns. In addition other natural resources must have been exploited including woodland, wetland, marine and mineral (iron, copper, gold?).



1.  The environmental assemblage, including the extensive animal bone collection, from Y Glŷn, Llanbedrgoch is of immense potential for understanding the economy and the range of activities on the site.



1.  Apart from the potential of Y Glŷn, it is impossible to identify any real strengths in our understanding of the economy, land use and the exploitation of resources; we currently know almost nothing, in part because of the difficulty of recognising settlement types.



1.  A programme of pollen cores throughout Wales would reveal major changes in vegetation over time, including the early medieval period. This has been very productive in Ireland where pollen cores are beginning to indicate regional differences in the early medieval period. In NW Wales it would be particularly interesting to examine whether there is a decline in agriculture in the post-Roman period and to what extent there is an increase in agricultural activity as the climate improves towards the end of the period.


2.  If early medieval settlement sites can be identified, a study of the landscape in which they are located should illuminate the exploitation of resources.


3.  Environmental evidence such as bone, plant and shell remains has the potential to tell us an enormous amount about the early medieval way of life and the economy. However its survival is rare in NW Wales because of acidic soils. If sites are to be excavated extensively, priority should be to those with surviving environmental remains.


4.  The exploitation of mineral resources in NW Wales has up to now focussed on the prehistoric and Roman period exploitation of copper and iron and widespread exploitation from the Industrial Revolution onwards. There are hints of early medieval copper exploitation, for example early medieval radiocarbon dates from the Great Orme copper mines, but there have received little attention. The potential for early medieval exploitation needs to be properly recognised and more fully investigated in the future.


5.  Marine resources. In Ireland early medieval fish traps have been identified in rivers and loughs. Fish traps are known from NW Wales. Radiocarbon dating of wooden examples could reveal ones of early medieval date. In N. Scotland shellfish and fishbone middens have revealed the exploitation of marine resources. If such middens could be identified, dated and sampled in the NW, similar potential exists.




C.  Ecclesiastical and Burial Sites


The origins of Christianity in NW Wales lie in the Roman period, though no evidence has so far been found. During the 5th and 6th centuries the evidence of the inscribed stones suggests that Christianity expanded, at least amongst the upper echelons of society, and during the 6th century major monastic sites such as Bangor are likely to have been founded. Around the 8th(?) century there may have been a gradual shift from burial in kin cemeteries to burial associated with ecclesiastical sites which eventually became parish churches and chapels of ease; in addition some early cemeteries developed into local church sites. Some minor churches may have been connected with landed estates. Over the last few years there have been major advances enabling recognition and increased understanding of early medieval ecclesiastical and burial sites in NW Wales. However, with the exception of burial sites, very little archaeological excavation has taken place.



1.  The early medieval inscribed stones and stone sculpture provide a major source of evidence for the development of Christianity in NW Wales, as well as for determining cultural (Irish Sea, Continental, Viking) connections. The Christian hic iacit formula predominates on the inscribed stones in this area and there is only one ogam and Latin inscribed stone. However only 40% of the inscribed stones can be associated with modern church sites. The majority appear to be associated with isolated burials or undeveloped cemeteries. A considerable number of inscribed stones are associated with prehistoric ritual and burial monuments or Roman roads. There are fewer cross-carved stones in this area than in the SW but these are mainly connected with church sites where they probably functioned as grave markers; a small proportion may have been way markers or have indicated boundaries or the ownership of land. The clusters of more ambitious 9th-11th sculpture, mainly crosses, indicate the most important sites, such as Penmon, most of which are also mentioned in the documentary sources.


2.  A considerable number of potential early medieval cemeteries, mainly undeveloped, have been identified in NW Wales, especially Anglesey, because of the use of long cists. Our knowledge of such burial sites has recently been reviewed by Gwynedd Archaeological Trust for Cadw. Over the last 30 years there have also been major excavations at Arfryn, Capel Eithin, Tŷ Mawr Holyhead and now Capel Ffraid, Treadddur Bay, where skeletal material is being recovered. This rarely survives in the acidic soils of NW Wales.


3.  The recently completed survey of early medieval ecclesiastical sites in NW Wales by Gwynedd Archaeological Trust for Cadw has, amongst other things, demonstrated the nature of the resource and the types of evidence that may be used to identify such sites. A hierarchy of sites can be detected ranging from major regional churches, such as Clynnog Fawr and Tywyn, to hermitages, small local churches, chapels and cemeteries.



1.  The lack of any major excavation of a church site of any kind in NW Wales.



1.  So far our knowledge of the early medieval Welsh church has been established on the basis of documentary and place-name evidence, early medieval inscribed stones and stone sculpture, the recognition of curvilinear churchyards and burials and occasional other evidence. With the exception of cemeteries very little excavation has been undertaken. This has to be the next step in order to increase our understanding of the archaeology of the early medieval church in NW Wales which will ultimately enable comparison with other areas of Wales and beyond, including sites in Ireland, for example Reask, Armagh and Clonmacnoise, and Scotland, for example Whithorn, Portmahomack and Iona, where there has been much more archaeological investigation over the last 30 years or so. There have been similarly major excavations on Anglo-Saxon sites in England, for example Jarrow, Barton-on-Humber, Winchester. (It should be noted that elsewhere there has been a tendency to concentrate on known important sites.)


2. Over the next decade or so a range of ecclesiastical sites in the NW with archaeological potential need to be investigated in more detail, including their surroundings and their resources, initially using non-destructive means such as aerial photography, geophysical survey and map evidence. A small proportion should then be targeted to identify potentially rich stratified datable contexts, preferably with surviving environmental evidence and human bone. When the results of this have been analysed the opportunity should then be taken to conduct one or two contrasting major research excavations focusing on the sites identified. These are likely to be abandoned sites, sites where there is a clear shift in focus or churches which become redundant or where there are major repair works being carried out. The abandoned church at Llanfihangel Ysgeifiog (Ang.) remains an obvious target.


3.  Curvilinear churchyards are usually regarded as early medieval, though only one example in NW Wales, Llangian, has been radiocarbon dated (cal. AD 430-670, 2 sigma). Opportunities should be taken where appropriate to section curvilinear earthwork boundaries in order to recover dating evidence.



1.  The early medieval inscribed stones and stone sculpture continue to be threatened by weathering, neglect and sometimes lack of governmental protection. It is therefore essential that Cadw establish a committee that will ensure a proper framework for the future protection and display of these monuments.


2.  The lack of major funding for archaeological investigation of church sites which are undergoing repair or have been made redundant means that the archaeological resource is under threat.




Paper prepared by Nancy Edwards (University of Wales Bangor)




This document’s copyright is held by contributors and sponsors of the Research Framework for the Archaeology of Wales.