A Research Framework for the Archaeology of Wales

Northwest Wales - Roman






Interest in our region’s archaeological remains, at both professional and amateur level, has been of long duration and is undiminished.  Nowhere is this better exemplified than in the Romano-British period, since NW Wales contains some of the defining type-sites of the age  -the enclosed and unenclosed hut-groups- together with some of the best known and extensively excavated military installations; Segontium being a prime example.  It is also one of the few British regions which Classical sources specifically mention: the Anglesey campaign of AD 60 and the final capture of the island by Cn. Julius Agricola in AD 77 being notable geographical markers in the context of first century military operations.  At the close of the Roman age it briefly figures in the context of the politico-military struggles of the fourth century  -the withdrawal of the Seguntienses- whilst its Roman past is also alluded to in the hero-tales and historical writings of the Early Christian period, in which Constantine the Great and Magnus Maximus figure large. 


For many years Gwynedd has exemplified ‘Romanization at the fringe’; an upland region devoid of those manifest indices of Romanity seen, for example, in SE, and to a lesser extent in SW Wales.  Following its conquest in the 70s of the first century and a garrisoning phase which effectively only lasted to the mid-second century, its native communities accepted Roman rule, but seemingly played no part in that self-motivating process that we term ‘Romanization’. Urban centres are notably absent, whilst there is no hint of villa development.  Military remains are all too obvious, and, in the case of Segontium, indicative of a long-lasting, but markedly localised garrisoning phase.  There is evidence for the exploitation of the region’s metals by Mediterranean-style business partnerships (societates), whilst Romanization has been perceived as being essentially limited to the circulation of a range of consumer durables  -pottery, coinage, metalwork and objects of worked stone, such as rotary querns- on settlement sites whose origins lie within the latter centuries BC.  When the politico-military infrastructure of Roman Britain collapsed in the early fifth century, there was apparently a return a pre-Roman socio-economic framework within the region.


Is this, simplistic, traditional view acceptable, especially in the context of publications such as Simon James and Martin Millett’s edited volume, Britons and Romans: advancing an archaeological agenda CBA, 2001), which exemplify new concepts /approaches to the study of the period based upon the application of archaeological theory, some of which, admittedly, are specific to a southern English landscape, others, such as ethnicity, which have an Empire-wide applicability? Surely, such questions cannot be addressed unless we have an opportunity to assess the traditional narratives, and then revisit them in the light of what we know, or, more importantly, don’t know.  Only then will be able to progress along these theoretical lines.


My brief is, firstly, an assessment of our present state of knowledge, based upon a SWOT analysis of what the Resource Audit contains, addressing the themes of Invasion and Military Occupation, Communication Systems, Rural Settlement, Technology and Trade, and, finally Ritual and Funerary activities. Secondly, to consider research priorities and strategies for the furtherance of our knowledge.  My comments incorporate those voiced by other members of the Romano-British working group, Richard Kelly and George Smith in particular, and those made at the Bangor seminar.



Invasion and Military Occupation



The conquest and pacification of the region in the early Flavian period is well established, and it is probable that the majority of the auxiliary forts founded in the train of conquest are known to us. These may be contextualised with reference to contemporary establishments in mid- and south Wales.


We have a reasonable overall chronology for the majority of the sites investigated, and also a good publication record.  The military presence diminished rapidly about AD 120-30.  Only three forts can claim some evidence of occupation to circa. AD 140-50, and only one certainly  -Segontium- thereafter.


We have an exceedingly rare overall plan of an earth and timber fort at Pen Llys Tyn based upon rescue work in the late 50s and early 60s, and basic plans of the interiors of stone forts at Caerhun (long a type-site for a quingenary unit), Pennal (based upon aerial photography and geophysical survey) and a multi-period interior at Segontium.  Our understanding of the structural sequence at sites such as Bryn y Gefeiliau has also been clarified as a result of geophysical survey undertaken by Gwynedd Archaeological Trust as part of the Roman Forts Environs Project.  The remarkable plan of the apparently single-phase, large earth and timber fort at Llanfor is testimony to the effectiveness of this technique.


At Tomen y Mur we have a truly remarkable survival of a military landscape, with a two-phase auxiliary fort, vicus, ?parade-ground, bath-house, ?mansio, leat/aqueduct system, military cemeteries, marching-camps, roads and the largest concentration of practice camps in Britain after the group on Llandrindod Common (Rad.).  This is a truly invaluable resource and still has much to contribute to our understanding of the Roman army of the Flavian-Hadrianic period and its impact upon the surrounding area.  The wetland nature of the fort’s surroundings holds the promise of high quality palaeoenvironmental data.


At Llanfor we have a unique complex of two marching-camps, a large earth and timber fort and a pentagonal enclosure, possibly a ‘stores-base’ or a ‘construction camp’ belonging to a campaigning phase, all as yet undated.


In Segontium we have the only continuously occupied, long-lived auxiliary fort in Wales, with occupation only terminating circa AD 393/4.  There is some evidence that the site served not only as a military base, but also as an administrative focus for the region from the mid-Antonine period to sometime in the mid-third century.


The late Roman politico-military situation in the region appears to have been radically different to that which prevailed over most other parts of Wales.  It appears to have been threatened with raiders/would be settlers from across the Irish Sea.  We thus have evidence for an integrated system of defence, certainly functioning in the early 390s, based upon a fort for limitanei at Segontium, a small coastal fort, presumably for a naval squadron at Caer Gybi, and a watchtower, the ‘eyes’ of the fort, on Holyhead Mountain.  The coastal warning system, if such it is, may have been more extensive.


The Roman Forts Environs Project has produced remarkable detail not only of the  planning of forts at Llanfor, Pennal and Bryn y Gefeiliau, but more importantly much of the layout of the vici associated with them. Such detail is unparalleled elsewhere in Wales. Here we have an excellent resource not only for the management of these complexes, but also for future research.


We have a reasonable understanding of the artefactual range present on military sites, and on the probable relationship between the late Roman garrison of Segontium and their suppliers, at both local and provincial level. 



Though a number of marching-camps belonging to the campaigning phase are known, they are too few and widely scattered to form a cohesive and comprehensible pattern in which relationships between camps and operational bases become evident.  Clearly, in the light of what we know historically of both pre-Flavian and Flavian operations in the region, many camps remain to be discovered.  The chronology of those known to us is unclear.  Only where they lie in close proximity to Flavian and later forts may we hazard a guess at their relationship and hazard an approximate date.


Short-lived semi-permanent military establishments associated with operations in AD 60, such as the fort(s) on Anglesey, and the base for that operation on the mainland are still undetected.


Key components of the campaigning phase, such as the large fort and ‘stores base’ at Llanfor are unexplored and their chronology, as well as the specific function of the latter, remains uncertain. 


All of the auxiliary forts in the region have been subjected to archaeological investigation, but the work has been frequently small-scale, and mostly undertaken in the first half of the 20th century.  This frequently revealed sketchy building plans belonging to the ‘stone’ phases and little else. It is clear that most forts have complex histories raising issues that cannot be addressed by small-scale work.  Even the date of the provision of stone defences and stone internal buildings is far from clear.  Pennal has not even been subject to excavation, nor has the fortlet at Brithdir.  Were these forts ever fully garrisoned throughout their occupation?  The Segontium evidence is instructive here, since it showed that garrisons could be reduced to diminutive size without a reduction in the size of the defensive perimeter.  Were units, then, commonly split between several forts?  Was Caerhun occupied beyond the mid-second century?  Is there a likely relationship between mansiones and later Roman material on fort sites?


Though we may be confident in Caer Gybi’s late Roman date, we have no specific evidence as such.  When was it built?  We also only possess a terminus ante quem for the associated watchtower on Holyhead Mountain.


The Roman Forts Environs Project has dramatically enhanced our knowledge of the extent and planning of military vici, but that is about all. Frankly, we know next to nothing even about their basic chronology.  The built-over vicus at Segontium produced material of Flavian-Antonine date, yet the fort was much more long-lived.  Was there a late vicus here?  Was the extra-mural settlement at Brithdir, for example, a small vicus or a works-complex?  How long-lived were the vici ?  Did any survive the abandonment of the parent fort? Who exactly lived there?  Did they have a role as market centres? These are big issues, but the dearth of basic information pertaining to their establishment, growth/contraction, abandonment and socio-economic role severely hinders any assessment of the relationship between the Roman army and the civilian community at large.


Communication Systems



The strategic road system, intimately connected with the needs of military supply, has been extensively researched, with aerial reconnaissance having proved particularly successful in the elucidation of the course taken by the road network to the SE of Tomen y Mur, to the E of Caer Gai, and again to the E and S of Brithdir.  Some fine stretches of road survive within the region: particularly significant are those whose course is largely indicated by quarry pits for road material.


There is still considerable interest in resolving the problem of road alignments at both professional and amateur level.  The input of the RCAHM and the Snowdonia National Parks has been notable.


There is, as one might expect, evidence for maritime communications as evidenced by the distribution pattern of ceramic containers as well as rare discoveries such as the Porth Felen lead anchor stock, which, though probably of the first century BC , could manifestly have been used at a much later date.



Substantial stretches of the road system are still conjectural.  There are big question marks against the presumed course of the roads from Pennal to Brithdir, Segontium to Pen Llystyn and Pen Llystyn – Tomen y Mur.  Similarly the route linking Bryn y Gefeiliau with Segontium, if such existed.  Nothing is known of the road system on Anglesey.


Virtually nothing is known of bridges on these routes.  The small bridge abutment south of Tomen y Mur is unique.


Very little is known of sea-borne communications, the existence of which is indicated by the presence of bulk containers such as amphorae on fort sites, and the distribution of ceramic products such as SEDBB1.



Rural Settlement



The region has long been known to possess one of the defining features of the RB upland landscape; stone-built settlements, typified by the enclosed hut group, and associated field systems.  These are numerous and widespread with complete landscapes surviving almost intact around the N and W margins of Snowdonia, less frequently on Anglesey. They exhibit a variety of forms, with indications of regional styles of construction


Excavations spanning three quarters of a century have produced a considerable body of evidence for the origins, development, and socio-economic background as well as the environmental context of some of these settlements.  Results from those investigated in the Graeanog area, circa 20km to the SW of Caernarfon have been particularly good. The stone-built settlements are now known to have origins in the latter part of the first millennium BC, to have complex histories and for some to continue in occupation to the mid first millennium AD at the earliest.  Not all are strictly RB in date as hitherto thought, though those with elements of straight, rather than curvilinear enclosing walls, and a mix of rectangular and circular buildings seem to consistently produce evidence of RB occupation, if not outright origin in the period.


Recent work has demonstrated that other types of RB settlement also exist: typically unenclosed, clay-walled round-house settlements, such as Bush Farm (Caerns.), Melin y Plas and Cefn Cwmwd on Anglesey, and rectangular, embanked and ditched farmsteads of the Bryn Eryr (Ang.) type.  The recognition that the rectangular enclosure, otherwise known throughout Wales and the Marches, is also a feature of the later prehistoric and RB landscape of the region, though not apparently in great numbers, has been an important discovery.


Continuity of site, and in some cases clear evidence of near continuous occupation from the late pre-Roman Iron Age into the Early Christian era is evidenced at a number of rural settlements, such as Pant y Saer and Cefn Cwmwd on Anglesey.

Indeed, by virtue of early medieval documentary sources, as well as defined type-sites, the region has an extremely important potential for research into settlement continuity and land-use.


The publication record in respect of the investigation of RB rural settlement in the region is good.


Many settlements exist in a clearly defined landscape context of terraced fields and enclosures.  These relict elements of an agricultural landscape, coupled with frequent areas of mire, enable the investigation of landscape/settlement history over the span of over a millennium, through the examination of associated soils/sediments, pollen spectra and macrobotanical remains.  We probably know more about the settlement history and landscape use of this region in the pre-Roman Iron Age and RB period than any other part of Wales.


A high percentage of hill-forts in the region produce evidence of activity within in the RB period, though the context is frequently unclear.  Some sites, such as the now vanished Braich y Dinas produce a range of artefactual evidence which cannot be readily matched at the farmstead sites, whilst others, such as Moel Trigarn, apparently show evidence of maintenance in the RB period.



The problem of the partial survival, and the relict character of the evidence, is paramount.  The stone-built settlements are grossly under-represented in the low-lying areas; antiquarian, particularly 19th century evidence for their destruction being important.  The recognition of timber/clay-walled settlements in these regions is also significant.   Excavations on the course of the A55 on Anglesey has for the first time provided us with some idea of the true settlement density in this part of the region.  It is manifest that many more settlements must have existed than are presently visible.


Research has tended to concentrate on the enclosed hut groups.  Very little work has been done on the isolated, upland-type stone huts, and the so-called ‘villages’ or agglomerations of huts.  We need to determine the relationship between the enclosed hut-groups, the ‘villages’ and the upland varieties of settlement.


Excavation has also been biased within the region.  The Crawcwellt area excepted, Merioneth has only seen small-scale little work, and until of late relatively little excavation has been conducted on Anglesey.


The isolated, upland huts remain a problem.  A few have produce RB material.  What is their origin and function, and how do they relate to the enclosed and other types of settlement?


The social context of rural settlement as a whole is still a matter of dispute. Some interesting points emerge in the context of the possible relationship between the farms on the Graeanog Ridge, but require testing against data from other groups of like monuments.  We are still largely ignorant of the way in which society was organised prior to the Roman conquest and how it was affected by it.


We do not know whether the settlement history of the region was static.  Did any new types appear?  What of the record of tile, samian and coarse-wares at Maenhir/Tre Anna?  Is this a Romanised rural dwelling, or perhaps the centre of an Imperial estate?  How many settlements transcend the conquest?  What is the settlement pattern over time?  What proportion continued in occupation into the post-Roman period?


Our knowledge of land-use in general is poor.  What proportion was under cultivation at different times?  How much woodland survived, and what economic use was made of it?


What is the relationship, at an economic level, between the differing types of rural settlement and the military?  The Segontium palaeoenvironmental data is a pointer to such links, but is it representative?  A comparative exercise is urgently needed.  In this respect the non-survival of animal bone on some the recently excavated hut-groups is a problem, since it severely limits inferences as to their economic base and overall relationships.  The excavation of settlements on limestone should, thus, by accorded high priority.



Trade and Industry



 The Roman period saw the introduction of new technologies and crafts, and, significantly, a fundamental change in the scale of production and exchange. Coinage, ceramics and other consumer goods were introduced into the region.  The military produced their own brick and tile, whilst entrepreneurs set up pottery kilns in the vicinity of forts catering for much of the garrisons’ requirements in terms of coarse wares from the 70s to the early second century AD.  Some of these local products reached native communities, to judge by the evidence from farms on the Graeanog Ridge.


The location and structure of military kilns for the production of tiles/bricks are known, and two - Dolbelydr and Pen y Stryd (Mer.)- may be linked, spatially and chronologically, with the fort at Tomen y Mur.


We have data on the currency of ceramic types in both military and civilian usage from the Flavian period to the close of the fourth century.  This represents an important contribution to the study of trade patterns in general, since ceramics in particular represents a ‘tracer’ commodity.


The incidence of coin use, including hoarding, is well established, and may be interpreted in the light of data pertaining to the coinage of Roman Britain as a whole.


Advances in water technology are demonstrable by the ubiquity of wells in fort/vicus contexts, leats (and by implication a timber aqueduct) at Tomen y Mur, and the epigraphic record of an aqueduct at Segontium.


Research by Peter Crew has demonstrated that regional iron production was well advanced in the pre-Roman period; may possibly have collapsed at the conquest, and, following a hiatus, was apparently allowed to continue at Bryn y Castell.  Whilst there is frequent evidence for smithing at enclosed hut-groups, we are uncertain as to whether this depended upon the continued exploitation of locally available iron, or upon a trade in iron billet.


We have explicit evidence for an extensive copper-extracting industry, apparently not in native hands but based upon societates (partnerships), a number of which are known from stamps on copper ingots.  There is circumstantial evidence for copper mining at Parys Mountain (Ang.) and the Great Orme.  There is also a little evidence for associated mining settlements; as at Tremadoc, though uncertainty remains as to its specifics.



Whilst it is clear that pottery kilns existed in the vicinity of most, if not all, early forts; their location is unknown, and the attribution of pottery fabrics to a regional rather than specific source is the norm. The spectrographic analysis of these local types may possibly allow us to tie them down to specific sources within the region.


The precise sourcing of the extracted metals is unknown, whilst the identification of the associated mining and smelting sites (?Tremadoc excepted) is yet to be determined.


Was the army involved in the exploitation of the region’s metals?


 Nothing is known of the chronology of the copper and lead/silver industry in the region, whilst the scale of production of these, and the iron industry, is presently impossible to quantify. Was iron being brought into the region in the form of billet, or was local, ?native production sufficient to meet demand?


Whilst it is clear that the region’s lithic resources  -slate in particular-   were being exploited, and distributed as far east as Cheshire and south as Breconshire, we know nothing as to the location of Roman quarrying or as to who was responsible for its operation.


We have virtually no data on some aspects of native industry/craft production, such as bronze casting, bone and woodworking, or dyeing and tanning?  Was some craft production centralized, or not?  Could palaeoenvironmental deposits on settlement sites throw light on these problems?


We urgently need to know what access different social groups, or similar groups spread throughout the region, had to the market.  The analysis of artefact assemblages will throw some light on this issue, though the scarcity of material from some areas will make such an analysis woefully incomplete unless more excavated evidence becomes available for study.


Burial and Ritual



The basic characteristics of intrusive ‘Roman’ funerary practices  -at least those associated with forts and vici of the late second to mid-third century, when cremation was the dominant rite- is well established.    Cremation burials are known from the vicinity of several forts, such as Segontium, Caerhun and Caer Gai, and some of the remains are curated in museums.


The region has examples of the exceedingly rare survival of funerary monuments above ground.  These are constrained to the vicinity of roads leading to the fort at Tomen y Mur, and comprise large, rectilinear, ditched barrows to the south and a linear arrangement of small, mounded burials to the NE.  These comprise a rare body of evidence for differences in burial rite in military contexts. Aerial photographs and geophysical survey also hint at the existence of stone-built funerary monuments close to the fort at Pennal.  Such instances are again exceedingly rare in Wales.


We have precious literary evidence for the existence of native cult sites destroyed by the army following the capture of Anglesey in AD 60.


The region has produced rare examples of dedications to, and sculptural representations of ‘Roman’ gods: Mars militaris, Minerva, Mercury, Mithras and a horned deity, possibly British in origin, from Segontium; whilst a fragmentary sculpture from Caer Gai either depicts Bacchus, or more likely Hercules.


The temple of Mithras at Segontium is the only published example of a focus for military cult observance in Wales, though the relief from Caer Gai alludes to another shrine.



The evidence for funerary practices is overwhelmingly biased towards military establishments, and even then towards the late first to mid- third century. With inhumation universal by the fourth century there ought to be inhumation cemeteries in the vicinity of Segontium, though none are known.


Epitaphs of serving soldiers and their families are known from north Britain in the second-third centuries, but, curiously, none from Segontium, though such ought to exist.


We are sadly ignorant of native funerary practices throughout the period.  There are no examples of burials from the settlements of this age, though the peripheries of such have not been tested.  The only certain example that we have of a native burial is that of Camuloris, an inhumation in a decorated lead coffin, probably of the mid-late fourth century from Rhuddgaer on Anglesey.


Encisted inhumations are recorded at a number of hut-group sites dug into in the nineteenth century, but it is unclear whether these are of the Early Christian period or earlier.


Other than dedications, shrines and temples associated with the Roman army there are no examples of built shrines, or for that matter ritual deposits, pertaining to native religious observances in this period.


We know nothing about the Early Christian community, or even if such existed in the region in a Roman context.



Opportunities and Research Priorities


Campaigning and Military Occupation

An obvious priority must be the search for the short-lived pre-Flavian sites that must exist in the region.  The garrison base(s) on Anglesey may be amenable to detection through aerial survey, as may marching-camps, which antedate the crossing of the Menai strait.


Similarly, the marching-camps for which a Flavian date may be suggested are far too few in number, and aerial reconnaissance may prove crucial in the detection of missing sites.  A further spin-off may be a substantial increase in the number of practice-camps in the vicinity of those forts where hitherto only a few such sites are known.  The dramatic increase in the number of practice-camps associated with Tomen y Mur within the past decade is instructive. The budget for such research-oriented flying needs to be substantially increased, with extra funding being available for those seasons when a bumper crop of new sites may be anticipated.


To continue aerial reconnaissance within a 5-7km radius of auxiliary forts in order to determine whether more examples of practice-camps are to be found.


A resolution of the chronology of hitherto undated military complexes, such as those at Llanfor, through research excavations.


Excavation is necessary to clarify the occupation sequences at those auxiliary forts for which we have the most imperfect knowledge, especially at sites such as Bryn y Gefeiliau where geophysical survey suggests a much more complex history.  At the majority the determination of the strength of the second century or later garrisons is a priority.  We may also be able to determine whether ancillary structures such as mansiones with their associated artefactual  assemblages are responsible for the apparent later Roman occupation sequence at some forts; for example, Caerhun.


Research excavations on military posts are vitally necessary in order to produce a suite of palaeoenvironmental data for the investigation of the relationship between the garrison base and its hinterland. In this respect a wetland site such as Tomen y Mur and its associated vicus have a tremendous potential for the survival of organic remains, which could be utilised to tackle the kinds of issues raised above.


It is important to determine the chronology of late Roman installations such as Caer Gybi, and also search for potential late Roman watchtowers on Anglesey and the mainland.


To complete the Roman Forts Environs Project, which has produced invaluable results.  The results have clarified the internal arrangement of some forts, whilst the plans of the vici not only allow the effective management of the forts’ environs, but also throw up new questions in respect of their interpretation.  The detection of pottery kilns is a potential spin-off of this project, as is the location of areas set aside for burial.


It is essential to undertake excavations at selected vici with a view to addressing their overall chronology, and their socio-economic role as an interface between the military and the native community.


The analysis of artifactual material from forts, vici and rural settlements in order to address broad economic perspectives: the operation of a market economy, barter systems, resistance to change etc.  The economic dimension, an integral part of the impact of conquest, will be difficult to grapple unless such an analysis is initiated.



Our knowledge base is exceedingly weak, and a substantial portion of the road network is conjectural.  Nothing is known of bridges, or alterations to the road network as is hinted at by apparently different alignments near Brithdir.  Research into this vital element in respect of military supply and a means of economic development is important.


There is still considerable interest in the investigation of road systems, and is a fruitful field for co-operation between professionals and amateurs.  The contribution of aerial reconnaissance in the detection of features such as quarry-pits for road material, in the absence of upstanding evidence for the roads, is vital. 


Close watch needs to be kept on the likely sites of river crossings in the RB period, together with a monitoring of possible anchorages/docking facilities, as, for example, close to Caerhun.  Native coastal traffic may possibly be revealed through concentrations of RB material.


Rural Settlement

Many of the remains are in a fragile state, and their preservation and record must be a high priority. The threat to the data-base through damage by stone-clearance or ploughing needs to be addressed by providing better protection through the increased use of scheduling, coupled with raising awareness among landowners/farmers of the importance of the sites on their land.


The reliability of data in respect of the overall distribution and density of the settlement pattern needs to be addressed, since the evidence, such as that gleaned from the A55 improvement scheme, indicates that large numbers of settlement sites have been lost through stone-robbing and agricultural operations. There is, thus, a need to continue field survey and aerial reconnaissance targeting those areas those areas where evidence for settlement, including, but not focussing exclusively upon hut-groups and field systems, is patchy or very thin.


Only a properly constituted research project, investigating selected types of rural settlement within different parts of the region, can answer some of the questions posed in respect of the chronology and socio-economic relationships of different components of the RB settlement pattern.  For example, so-called ‘villages’; isolated, upland huts, and the better known enclosed hut-groups.


Similarly, sites where Romanised buildings are probable, such as Maenhir/Tre Anna, require investigation to determine whether these are high-status settlements within a rural settlement pattern, or have an altogether different function.


It is essential that research programmes embracing later prehistory take the RB dimension into account, and that collaborative schemes be undertaken.


It is important that the investigation of settlements on limestone is accorded a high priority, in order to address issues arising from the poor survival of animal bone on sites hitherto excavated.


A high priority must be given to the investigation of landscape characteristics from later prehistory to the Early Christian period.


We still have minimal detail concerning the role of fortified sites in the settlement pattern and socio-economic structures of the age.  Very little work has been done on the hill-forts of the region since the RCAHMW’s surveys in the 1950s and early ‘60s.

The effects of the conquest might be recognised through abandonment or change of use. Opportunities for the investigation of these sites must be accorded a high priority.


Trade and Industry

Continued academic interest in artefact assemblages and the distribution of finished goods/consumables will lead to the analyses of material within the region.  We will thus be in a better position to discuss issues such as availability, selection and the social meaning of RB artefacts on a wider front.


The location of kilns for the production of pottery in the vicinity of auxiliary forts should be accorded a high priority.


There is a need to locate metal-extraction and quarrying sites, and to determine whether these have a settlement component, as in NE Wales.  The chronology and scale of such exploitation needs to be addressed.


‘Domestic’ industrial processes and craft activity at vici and rural settlement sites demand urgent investigation.  What is its history and scale? Is there specialisation at such levels of production?


Burial and Ritual

High priority must be given to the detection and excavation of burial sites, either by the use of geophysics in the vicinity of settlements, or by the investigation of multi-period burial sites (Bronze Age – Early Christian), which may throw light on one aspect of native funerary practice.


The detection of RB cult sites in either vicus or rural contexts is a high priority.  The investigation of the interiors of hill-forts, which have produced RB material, may prove fruitful in respect of the latter.




Paper prepared by Jeffrey L. Davies (University of Wales, Aberystwyth)



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