A Research Framework for the Archaeology of Wales

Southeast Wales – Later Prehistory





● Our aim is to pose relevant and interesting questions, in order to stimulate research leading to an improved understanding of later prehistory in south east Wales. It is important that these are suitably phrased and feed into contemporary ‘big issues’ within academic research, so that studies from this part of Wales might be made to illustrate wider themes and processes.


[Here, we have chosen to focus upon content and research approach, keeping the scope broad and open. Delivery mechanisms for specific projects will develop out of wider consultation, once this paper has been discussed. We recognise that the outcomes and findings of other period and regional groups are likely to have important bearing on which projects emerge as being particularly resonant and timely.]


● Developer led projects need to maintain the aspiration to contribute to major research questions and to achieve the highest standards. In relation to certain high profile developments, some developers are prepared to pay for the very highest quality flagship projects. Many, on the other hand, will require a more basic level of work so that development may proceed.


● Inevitably, much development led work is reactive, small-scale, uneven in coverage and may be carried out in a fragmented way. The reality is that developer led work will only occasionally be able to provide a complex and subtle understanding of past lifeways.  In many cases, the full potential of developer led research may only be achieved if it is linked to wider research programmes.


● Therefore, it is very important that all archaeological bodies working in Wales (e.g. Cadw, RCAHMW, NMGW, the Welsh Trusts and other archaeological contractors, University Archaeology Departments, National Park Archaeologists and voluntary groups), each with their own strengths and emphases, work together to formulate strategies for understanding the Welsh archaeological record. A change in attitude towards research and its funding would be helpful. Co-ordinated research programmes relating to issues, areas and aspects of the archaeological record needing clarification should be identified.


● Given their present structure, geared to the needs of development control, SMR databases do not themselves provide the necessary information to either pose or answer many of the types of research questions, which are of current concern. Lists of database entries are no more or less useful than catalogues of books in a library, whilst distribution maps are only useful when they attempt to examine and answer particular research questions. Consideration needs to be given to how SMRs may be upgraded and maintained in the future to provide an increasingly effective research tool. This would lead to a better understanding and protection of our heritage.


● Following cross regional and cross-period study of the four Resource Audit Documents, wide variations in the numbers and proportions of records per type and period seem apparent. These often appear to have more to do with variations in Trust inputting and interests (within a constantly under-funded environment), than with relative archaeological abundance or scarcity within any given region.




(What is exceptional or noteworthy about the known later prehistory of south east Wales?)


● The diversity of hillfort forms within the region (promontory coastal, inland promontory, wide spaced multivallate, hill slope, contour, cross-banks, annexes, univallate, bivallate, multivallate, small to medium size range), as opposed to their relatively modest frequency (125) by comparison with other regions of Wales.


● The abundance and importance of Late Bronze Age metal hoards, in contrast with the dearth of corresponding settlement evidence and secondly, with comparable evidence from other regions of Wales (45 LBA hoards, Llantwit-Stogursey tradition, Llyn Fawr phase and hoard).


● The survival of a well preserved alluviated coastal landscape along the Gwent Levels, revealing an abundance of structures, including Middle Bronze Age to Iron Age settlements some with a rectangular architectural tradition. This contrasts with the roundhouse tradition in this period in the rest of Wales and England.


● In contrast with many other parts of Wales, the material culture record from existing Middle and Late Iron Age settlement and hillfort excavations is reasonably visible and diagnostic, (e.g. ceramics, iron and bronze artefacts, glass, coins), although this potential remains to be fully explored. Hoards and single finds complement this material record.  They include a now sizeable group of artefacts displaying the late La Tčne art style and relating to personal adornment, horse display, feasting and rich burial during the Late Iron Age.


● The region is made up of varied physical landscapes, providing a range of natural resources upon which to draw for material culture production and offering opportunities for varied settlements, landscape organisation and subsistence patterns.



● Virtually nothing is known about later prehistoric landscape organisation. The only tentatively dated evidence appearing within the audit document, are Bronze Age clearance cairns. Few field systems or boundaries of later prehistoric date are recorded.


● Very little is known about inhabited places. Little non-hillfort settlement is identified (29 sites). Comparatively few hillforts and settlements (each less than 20) have received any excavation, whilst very few have received recent or large area excavation.  Compare with 600 defended enclosures in west Wales; over 400 open settlements in northwest Wales, 234 enclosed settlements in northeast Wales.


● Our understanding is hindered through strong geographical and chronological contrasts in the visibility of archaeological settlements and landscape features and, secondly, in archaeological coverage.

-          The former county of Gwent has received little survey, fieldwork and synthesis with the exception of the Gwent Levels and immediate hinterland.

-          Visibility within upland areas of Neath-Port Talbot, Rhondda Cynon Taff, Merthyr Tydfil and Blaenau Gwent is poor (and results of the recent RCAHMW Uplands Survey work remains poorly known to a wider audience). 

-          Chronologically, the Middle Bronze Age to Early Iron Age is very poorly represented in all areas (with the sole exception of the Gwent Levels), by comparison with both earlier and later periods.


● The existing chronological framework for later prehistory is very weak. With the exceptions of recent dating research of outstanding importance along the Gwent Levels and at Caldicot, series of radiocarbon dates for excavated later prehistoric sites remain extremely rare. This is in marked contrast with many other parts of Britain, where dating programmes have for some time been accepted as routinely necessary. Known ceramic assemblages and diagnostic metalwork remain unsorted. Few material culture associations have been independently radiocarbon dated.


● The burial record remains sparse during the Middle Bronze Age to Late Iron Age, in common with many other parts of Britain. However, the few known burials hint at a diversity of minority burial and ancestor linked rites in the region. Few burials have been radiocarbon dated. We understand very little about religious and ritual expression more generally here (e.g. the continued use of Neolithic and Early Bronze Age ritual foci, ritual/structured deposits and temples).


● The large number of known hoards and single metalwork finds from the Middle Bronze Age to Conquest period remain poorly integrated within wider interpretations of settlement and society. These have received little research and reinterpretation over the past twenty years as ideas and interpretations have changed elsewhere.


● Existing environmental evidence (pollen, plant macrofossils, charcoal, animal bone, molluscs, soil studies) for later prehistory is sparse and scrappy, when broken down to period and region. The sole exception to this pattern is the significant body of evidence now available from the Gwent Levels and the Lower Nedern Valley. The audit document includes only 22 Bronze Age and 34 Iron Age datasets from archaeological sites in southeast Wales and many of these were collected decades ago. Poor retrieval of these potential sources of evidence (even allowing for patchy preservation environments) are limiting our understanding of landscape character, farming and subsistence organisation and the social realms of eating.


● Few have attempted to write syntheses of the Bronze and Iron Ages of south east Wales. A mass of potential information remains unsorted:-


-          Developer funded archaeology, together with surveys for unitary authorities, Cadw and RCAHMW, are producing a large quantity of grey literature, which is poorly accessible to the wider community.

-          Museum collections (e.g. Merthyr Mawr Warren), AP archives and SMR databases are infrequently accessed and reinterpreted by archaeologists.

-          Artefact studies do not occupy the central role they deserve within our interpretations. Many discoveries have not, within the last twenty-five years, been incorporated within artefact corpora and therefore wider knowledge of their existence remains weak.

-          Key excavations and assemblages remain unpublished or are in the process of being prepared for publication (e.g. Twyn-y-Gaer (hillfort), Lodge Hill (hillfort), Trostrey (open/palisaded settlement), Greenmoor Arch (open settlement), Portskewett New School Site (enclosure), Rhoose Airport (enclosure), Burry Holms (roundhouse outside hillfort), Cae Summerhouse (hillfort), Caer Dynnaf (hillfort), Magor (settlement), Caerau (hillfort)).

-          The Portable Antiquities Recording Scheme for Wales is now generating new artefact and hoard discoveries, which deserve full publication and wider interpretation.




This section includes future research approaches and questions, which we feel would merit further consideration.


● Asking questions relating to the existing dataset for monument or artefact types of a given period. This would lead to a better appreciation of the diversity, development and character of the sites or material culture involved.



-          Sampling projects – small excavations of groups of particular monument forms or constructed landscape features (e.g. promontory forts, hillforts with wide spaced concentric ramparts, enclosed settlement forms, field banks and boundaries) could be undertaken in conjunction with a radiometric-dating programme.

-          A thorough synthesis of Middle and Late Bronze Age metalwork from the region (similar treatment of Early Bronze Age metalwork is also required), drawing upon new discoveries and the backlog of existing unpublished records. Integration with landscape and settlement evidence necessary.

-          A systematic survey of later prehistoric ceramics, in order to assess frequency, chronology, form, production etc. This would complement the existing Later Prehistoric Ceramic Survey data in England acting both as an effective research and planning tool.


● To focus (for an agreed timescale) upon a targeted geographical area/landscape block, employing a multi-period methodology involving environmental analyses, phosphate survey, aerial photographic survey and plotting, geophysics, radiometric dating and limited excavation. Examples might include some of the following:-


-          An upland area of Glamorgan or Gwent (exploring the apparent dearth of visible later prehistoric features using an integrated survey and excavation sampling methodology). 

-          The middle and lower Usk river valley (an area of suspected settlement density and diversity, yet to date poorly explored).

-          A transect from the Gwent Levels northwards to look at dryland evidence and its relationship to the wetland Gwent Levels evidence (if possible in an area of significant medium to long term development pressure to achieve linkage to developer funding – e.g. Llanwern and the south Newport motorway).

-          Part of the Vale of Glamorgan (following up recent survey work by the RCAHMW at Aberthaw and GGAT’s Roman Settlement in South East Wales Survey)

-          Gower (a candidate for identifying surviving prehistoric landscape evidence around known hillforts, following bracken clearance).


This approach would complement work undertaken on the Gwent Levels, which has been the most important later prehistoric project in the region over recent years. Such intensive effort would target areas where the existing archaeological coverage was poor, but of necessity would be geographically limited in coverage.


● Related, but worthy of separate note, we would encourage further efforts in GIS based landscape mapping, using all available sources of evidence, as a means of working towards an enhanced understanding of later prehistoric landscape organisation. Point and site oriented databases, on their own, fail to record and capture the relational complexities of geographically extensive archaeological evidence. This method would serve to aid the selection of areas for future intensive study.


-          One very basic question which requires some thought and demonstration is ‘where are the bounded later prehistoric landscapes in south east Wales?’ and conversely ‘where are the open landscapes in south east Wales?’ 

-          An important but simple task would be the aerial mapping of possible lynchet systems, particularly on the limestone areas of southeast Wales, where they should be well preserved if present.


● To look at artefact and resource based projects, seeking to clarify the availability of raw materials for artefact production. Clay, stone, salt and metal are important later prehistoric resources, relatively abundant in Wales but little systematic research is available on their exploitation. Analysis of sources, artefact composition and deposition should tell us much about exchange mechanisms, social relations and ideology.


-          If southeast Wales was inhabited by an artefact poor society, then how should we understand this ‘Celtic’ society when conceptions of the Celts held by archaeologists emphasise the importance of material culture and weaponry in particular?

-          What were the reasons for the apparent near abandonment of a well-developed Early Bronze Age ceramic technology during the Middle Bronze Age? Secondly, what social changes led here to the re-introduction of ceramics during the Middle to Late Iron Age?

-          How abundant is the evidence for the production and use of querns and what does this say about the organisation of societies here?


● Improved chronologies are a prerequisite and require a full campaign of radiometric dating, which encompasses all new projects, but also involves detailed research into museum collections and archives to find appropriate samples of known context to be dated. Particular emphasis should also be given to the dating of diagnostic metalwork and ceramic associations in order to test existing typologies.


● Even in a region of variable soil preservation conditions, a consistent and sustained commitment to sampling and research programmes for all forms of environmental evidence, would produce dramatic effect on our understanding of later prehistory. The importance of sieving strategies at appropriate sites to establish the economic basis and the extent of crop production needs emphasising.  This evidence only yields maximum potential when matched with high quality survey and excavation evidence (chronological resolution, understanding of soils and formation processes, full awareness of contextual associations).


Environmental studies require better integration within our interpretations of social life. Developer funded work needs to be integrated within wider research programmes, whilst methods of collection and quantification must be compatible and therefore comparable.


-     What was the balance between arable and livestock?

-          How were farming communities organised?

-          What is the evidence for agricultural surpluses and specialisation?

-          How mobile or sedentary were Middle and Later Bronze Age societies?

-          How densely occupied, cleared and managed were landscapes?

-          What part did upland and coastal environments play within wider agricultural regimes?

-          What social practices surrounded eating and feasting?

-          Were ideologies expressed through the placing of structured deposits of animals and plants?


● Specific questions:-


-          What is the character of the Middle Bronze Age in this region and how is it different from preceding and subsequent periods? For how long did ritual practices continue to be undertaken within earlier Neolithic and Early Bronze Age monumental landscapes?


-          When were hillforts and settlements first enclosed and defined? Are there palisaded enclosures or open settlements beneath hillforts?


-          How artefact poor is the Early and Middle Iron Age of south east Wales? If so, why?


-          How common were open settlements within later prehistory? Where and when and what they say about social relations.


-          What is the evidence for increasing social hierarchies, identities and economic specialisation during the Iron Age? When does this become visible by comparison with other regions of Britain?


-          What arenas of display are represented within the Late Iron Age material culture of this region? What does the settlement and material culture evidence of the Late Iron Age and Conquest period say about the character of interactions with the Roman world?


-          How influential were cultural ties with other neighbouring regions bordered by land and sea? How is this represented in the material culture of this part of Wales?


-          Are there Late Iron Age temples and shrines in south Wales?


● Historical literature for the early Medieval period exists in Wales and aspects of this literature have often been argued to provide information on social structure, customs and religious practices which persisted from pre-Roman times to the early Medieval period. Such inferences may frequently be problematic. The sources are, however, an important and distinctive aspect of the heritage of Wales and archaeologists must be prepared to work with historians, Celtic scholars and linguists to explore and debate the contribution that early historical sources can make to an understanding of the later prehistory of Wales.




We have taken the view to end this paper on a positive note and to reduce repetition by not listing the threats for later prehistory in south east Wales. We feel the dangers of assuming a status quo will be self-evident to readers.



Feedback during the regional day seminar (23/11/02)


Mike Walker: Was in accord with the need for improved retrieval of environmental evidence and radiocarbon dating programmes as routine. He informed the speaker of recent Cadw funded environmental work undertaken at the site of the Late Bronze Age hoard at Princetown.

Martin Bell: Re-iterated the importance of the survival of organic material culture on the Gwent Levels, in particular noting the tool marks preserved on worked timber.

Mike Hamilton: Was complementary about the content of the presentation and echoed concerns with the content of the SMRs and resource audit documents.

AG 26/03/03



Paper prepared by: Adam Gwilt (NMGW - chair), Martin Bell (University of Reading), Bernice Cardy (Swansea Museum), Mary Davis (NMGW), Mark Lodwick (NMGW), Graham Makepeace (University of Cardiff), Peter Northover (University of Oxford), Frank Olding (Blaenau Gwent, formerly Cadw), Niall Sharples (University of Cardiff), Adam Yates (formerly GGAT).




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