The story starts eighty years ago with two archaeologists, Cyril Fox (later Sir Cyril and one of the most eminent archaeologists of his day) and his wife Aileen, walking the upland ridges of Glamorgan whilst doing research for the County History that was then being prepared for publication. On Gelligaer Common they came across strange level platforms cut into the upper slopes of the ridge. Unsure what they were, Aileen Fox returned a few years later, in 1936, to excavate three or four of them, and discovered that there had been medieval houses erected on them.
A few other platforms were discovered in south Wales before the Second World War, but it was an article by Colin Gresham of the Royal Commission on the Ancient and Historical Monuments of Wales, working in Caernarvonshire, in 1956, that alerted archaeologists to the fact that there were platform houses in north Wales as well. It may be more than a coincidence that it was reportedly little more than four years later that Hen Caerwys was discovered, and that it was another member of the Royal Commission, Wilfred Hemp, who was instrumental in correctly interpreting the site as a medieval farming settlement. We’ll say more about this in the next instalment of our project diary in a couple of weeks’ time.
Geographically, it may seem a long distance from Glamorgan to Flintshire, and a generation separates those initial discoveries on Gelligaer Common from that of Hen Caerwys, but the identification of platforms provides a direct link between the two. For me there is also a personal link, for Aileen Fox was my teacher when I went to Exeter University in the late 1960s. She was at the end of her career then; I was at the beginning of mine, even if at the time I didn’t know it.
During the 1960s through to the 1980s a steady trickle of new platform sites were recognised in Wales but it was only when we started to pull all the information together during a Cadw-funded survey of deserted medieval rural settlements in the late 1990s that we began to realise how widely spread platforms sites were across Wales. The plan reproduced here was drawn up for our volume entitled Lost Farmsteads and published by the Council for British Archaeology in 2006 and shows the distribution of platforms in Wales. But you’ll notice that there are very few known examples in north-east Wales. This is what makes the house platforms at Hen Caerwys so interesting - they may be representative of a common enough settlement type in Wales, but in Flintshire they are a rarity.
Our understanding of most archaeological sites follows a fairly standard trajectory, not a set of rules, but more a consistent sequence in getting to know the site. First, of course, it is discovered and a record compiled, usually quite short, which is published as a note in a journal or entered into the regional archaeological record. Then the site is systematically recorded in a carefully measured survey, so that everyone can appreciate its overall plan and layout, as well as its geographical setting. Next, if there is sufficient interest, the site will be excavated, perhaps only a single trench, perhaps, if resources allow, a much more extensive examination. And finally the excavators will upon completion of their work, compile a report on what they have done and publish it, making the information available to all those who are interested.
A site plan prepared by Mr & Mrs Hill in 1983
In almost every single respect Hen Caerwys fails to conform to these norms.
The settlement is said to have been found by Wilfred Hemp, who until 1946 was the Secretary of the Royal Commission on the Ancient and Historical Monuments of Wales. In consultation with the eminent local historian, Ellis Davies, the site was attributed the name Hen Caerwys, implying that they felt it might be the predecessor of the medieval planned settlement lying little more than one kilometre away. But when did this happen? A date of around 1960 has been mooted, yet Hemp was nearly 80 by then, and the site wasn’t in fact scheduled for another twenty years.
However, by 1962 and seemingly before any standard preliminary work was undertaken, the newly named Hen Caerwys was being excavated. Two of the leading lights of the Flintshire Historical Society T. T. Pennant Williams and M Bevan-Evans, started to examine one of the house platforms, the work being continued by G. B. Leach and Pennant Williams in the following years. We know from short journal notes and the few surviving records that excavations continued until 1968 or perhaps even 1969, and that having completed the excavation of one house and its adjacent ancillary building, they moved on to another house platform nearby.
But Leach and Pennant Williams never properly published the results of their excavations – and indeed their brief notes in various journals became shorter as the years passed. It was left to T Rogers, who does seem to have been personally involved in the excavations, to publish House 1 in 1979. He had access to some plans and notes, mostly produced by Leach, as well as some of the finds, but virtually none of these are now available.
Rogers’ interest in Hen Caerwys coincided with Cadw’s predecessor, the Department of Public Building and Works’ realisation of the importance of the site, and in 1979 part of it was scheduled as a site of national importance.
It took thirty years before a comprehensive survey was completed. Under its woodland canopy Hen Caerwys represented quite a formidable challenge to the surveyor. Honourable early attempts by a Mr and Mrs Hill of Caerwys in 1983 (see illustration) and the Cadw Field Monument Warden Lorna Ackroyd Bell in the 1990s correctly identified and plotted some of the salient features of the site but were ultimately defeated by dense vegetation.
Only when Cadw grant-aided the Clwyd Archaeology Service to prepare a plan and report of the site, and the work was undertaken on the latter’s behalf by the Gwynedd Archaeological Trust in 1993, did a survey appear. This was made possible by the first large-scale clearance of the remains by the present owner Mike Owens as part of a long-term management plan, which over the years has gradually revealed the monument as we see it today.
An article from the Flintshire Leader newspaper, October 1964
In the last entry we mentioned several individuals by name who had been involved with Hen Caerwys in the past. We thought it would more be useful if we said a little more about the main characters. We are fortunate to be able to draw on details from the Flintshire Historical Society’s recently published Centenary History compiled by Bill Pritchard.
Wilfred Hemp who is credited with first recording Hen Caerwys was the secretary of the Royal Commission on the Ancient and Historical Monument of Wales from 1928 until his retirement in 1946. Hemp was responsible for the first of a new generation of Commission Inventories, on Anglesey, which appeared in 1937. When the Commission was evacuated during the Second World War he ran the organisation from his home in Criccieth, and he died there in 1962.
Hemp reputedly discussed the site with Canon Ellis Davies and between them they came up with the name Hen Caerwys. While Hemp’s links with Flintshire do not appear to have been particularly strong, Davies was a Flintshire man through and through. Born at Nannerch in 1872, he was appointed rector of Whitford in 1913 and stayed there until his retirement in 1951. He was one of those clergymen who had strong antiquarian interests, and is remembered best for his Prehistoric and Roman Remains of Denbighshire which was published in 1929, and the companion volume on Flintshire which appeared twenty years later, both of them still useful as sources of original information. He died at Bryn Derwen in Caerwys in the same year as Hemp, and admirable obituaries for both of them can be found in volume 20 of the Flintshire Historical Society Publications.
And so to the two main excavators of Hen Caerwys. Chairman of the Flintshire Historical Society from 1955 to 1968, T. T. Pennant Williams was a solicitor by profession, and an enthusiastic amateur archaeologist who could list work on the Roman site at Prestatyn amongst his many achievements. There is a personal link here too because he was the first legal advisor to my organisation, the Clwyd-Powys Archaeological Trust. Of G. B. Leach (d.1975) we have been able to discover less. He too was active in the county society and an enthusiastic excavator, working on the hillfort of Moel Hiraddug above Dyserth and on the castle site at Henblas near Bagillt, before turning his attention to Hen Caerwys.
Finally there is Tom Rogers, the most mysterious of the five Hen Caerwys celebrities. It seems that he was a Canadian, but little is known about his interest or involvement in archaeology until he appeared in north Wales, probably at some point in the 1970s. He had indeterminate links to the university college at Bangor, but seems to have had a particular interest in Hen Caerwys, publishing a short paper in the Bulletin of the Board of Celtic Studies in 1979 on Pennant Williams and Leach’s excavations and expressing an interest in conducting further excavations there. These plans came to nothing, and Rogers’ interest shifted to Arthur’s Cave in the Wye Valley near Symond’s Yat, where he claimed to have discovered Palaeolithic cave paintings or engravings. These claims were rapidly dismissed by several specialists in 1981 who viewed the marks as natural formations, while Rogers’ discovery in the cave of hearths and artefacts of Middle and Late Upper Palaeolithic date could not be verified. After that Tom Rogers fades from the archaeological picture.
A sketch prepared by Cadw's Field Monument Warden Lorna Achroyd Bell
Complex, in parts overgrown and difficult to fully appreciate on the ground, the low earthworks of Hen Caerwys contrast sharply with the more immediately recognisable castles, abbeys, industrial remains and prehistoric monuments in state care that are perhaps more readily associated with Cadw. These are widely publicised through an official website, leaflets, guide books and in the press; less well known are the roles of Cadw in the legal protection, conservation and management of the thousands of monuments across Wales that remain in private hands. Indeed, Hen Caerwys is one of the few deserted village sites in Wales that is openly accessible to the public and the clearly visible earthworks we see today are the result of almost 30 years interaction between Cadw (the Welsh Office before 1984) and two generations of the Owens family.
Scheduling Under the Ancient Monuments and Archaeological Areas Act of 1979, the Welsh Government is required to ‘compile and maintain a schedule of ancient monuments of national importance’. Monuments included in the schedule (of which there over 4000) are nearly all archaeological sites, ruins or buildings for which there is little prospect of economic use, distinguishing them from listed buildings. They range from Palaeolithic cave sites to ruined churches and WWII pillboxes. Scheduling does not affect the ownership of a monument nor does it give any right of public access but legally prohibits demolishing, destroying, damaging, removing, repairing, altering, adding to, flooding, or covering up a monument without first applying to Cadw for Scheduled Monument Consent. Metal detecting and archaeological excavation are included within these prohibited works. Recognised as a rare example of a deserted medieval village in north-east Wales, Hen Caerwys was first scheduled in 1979, the protected area being slightly expanded in 1980.
Field Monument Wardens Once a monument is scheduled, its condition is monitored on a five-yearly cycle by Cadw Field Monument Wardens (FMWs). These archaeologists effectively form the front line in monument conservation, recording the state of each monument in their region, reporting any damage or management issues and where possible, supported by the law, working with the owners to address these. The first FMW to deal with Hen Caerwys was Lorna Ackroyd Bell, who remembers playing amongst the spoil heaps of the original excavations, with which her father was involved. Lorna’s meticulous reports and sketch plans chart the progressively overgrown state of areas of the site, preventing her from producing an accurate plan, although her efforts (illustrated here) correctly identified many key features and would not be improved upon for almost a decade.
Cadw Management Agreements After Hen Caerwys was first scheduled and until the late 1980s Mr Owens Sr. participated in what was known as the ‘Acknowledgement Payment Scheme’, through which landowners received yearly payments to retain monuments on their land. Whilst some of these agreements also specified low-level maintenance works they were by no means systematic management plans. There is however, little doubt that the scheme was instrumental in preventing a number of sites from being ploughed or otherwise destroyed. Hen Caerwys was safe in the hands of an interested and sympathetic owner such as Mr Owens but it was conceded that he could hardly be expected to keep such an extensive area of woodland free of scrub and the FMW reports from the period chart the gradual disappearance of the site into the undergrowth.
By the early 1990s Acknowledgment Payments had been replaced by five-year Cadw funded Management Agreements, more actively geared towards conservation and tailored specifically to individual sites. In collaboration with a young Mike Owens who had assumed responsibility for the woodlands from his parents, Cadw funded a detailed plan to begin the clearance and long-term management of the remains in 1993, devised by André Berry of the former Clwyd Archaeology Service, an archaeologist with a background in ecology. This became a very successful Management Agreement which has been renewed several times, reflecting the ongoing effectiveness of Mike’s hard work. In order for an archaeological site to be effectively managed, an accurate record of the remains is required so with financial support from Cadw, Clwyd Archaeology Service commissioned a detailed topographic survey following extensive initial clearance of the site. The plan and information generated by this survey would also be used to form the basis of an on-site information panel and guide leaflets designed to accompany newly opened permissive paths through the woods, with which many of our readers will be familiar.
The object of managing a woodland site like Hen Caerwys is not simply to remove all vegetation from the archaeological remains but to maintain a balance between the monument and the continued exploitation of the environment in which its stands. Other Cadw management agreements are for sites under different types of land use such as grassland and moorland or even in gardens. The present appearance of the wooded portions of the site is the result of almost two decades of selective tree thinning and an extensive twice-yearly cutting of low-lying vegetation covering the key archaeological features. This is a carefully considered process, annual monitoring meetings with Cadw assessing progress and deciding upon the course of work for the following year. Such decisions can be specific to individual trees and bushes or to large areas of scrub, the need for long-term commitment being demonstrated by the gradual disappearance of some of the remains identified during the 1993 survey under regenerating vegetation. We will take a closer look at Mike’s woodland management regime in a later edition of the project diary.
Archaeology With Hen Caerwys stable, well-managed and accessible, Mike had discussed the possible origin and function of the various remains with successive Cadw staff, most notably the large rectangular enclosure in Coed Gerddi-Gleision. However, enquiries as to whether the site could be excavated were met with gentle refusals or suggestions that more survey work and background research was required. Indeed, archaeological excavation is a destructive process and for Cadw to grant consent for such work it needs to be convinced that the benefits of the project outweigh the potential damage to the monument. In addition, the Inspector requires proof of the competence and experience of the excavator, that the project is properly planned and resourced and that the proposed trenches are in acceptable locations.
Mike removed one of these obstacles early in 2011, taking the very unusual step for a private owner of commissioning the Poulton Research Project to carry out a resistivity survey before broaching the subject again. Later the same year, increasingly impressed by Mike’s commitment, Will Davies, Cadw’s Regional Inspector of Ancient Monuments was given a small budget to organise events for the CBA Festival of British Archaeology. Hen Caerwys, with its poorly documented 1960’s trenches, a fanatical landowner and apparently broad raft of local interest seemed an obvious contender for a community excavation. Clwyd-Powys Archaeological Trust, the regions foremost archaeological institution were approached to undertake this work in partnership with Cadw. Once this had been agreed, work began trawling archives and libraries and quizzing a variety of individuals in an attempt to locate the missing finds and records from the 1960’s excavation. In the meantime the new excavation was being arranged...
Hen Caerwys is a large and interesting site and we wouldn’t be doing it justice if we described it in one short section. So we’ll describe part of it here, and part in a section in two week’s time together with a summary.
Part of the plan originally prepared by the Gwynedd Archaeological Trust in 1994-95, and used in the Clwyd Archaeology Service guide leaflet
Traditionally regarded as a deserted medieval village, the remains scheduled as Hen Caerwys are better described as a tract of a relict landscape scattered with features of settlement, agriculture and industry of potentially many different periods. The ‘site’ occupies a limestone plateau, 1.5km to the NE of the planted medieval borough of Caerwys, the ground falling away to the south into a broad and shallow stream valley. Relatively modern enclosure roads and boundaries have divided the remains into three principal areas, the woodlands of Coed y Marian to the NW, Coed Gerddi-Gleision to the SW and the open pasture belonging to the farm of Marian Trefedwin to the NE. The remains also extend into a corner of Coed y Carreg to the SE, separated from the rest of the site by an earlier lane. With the exception of the fields belonging to Marian Trefedwin (to which there is no public access), the wandering earthworks and tumbled walls are very difficult to fully appreciate as a whole due to their wooded location; it should be stressed that most of the site formerly lay in open, common ground and will have continued beyond present property boundaries. The following description is neither a guided tour nor a comprehensive study of this extensive and complex monument but is intended to outline the principal features and their function. The site map reproduced here is that used on the former guide leaflets to Hen Caerwys, generated in 1994/5 from Gwynedd Archaeological Trust’s survey for the former Clwyd Archaeology Service. The features referred to in the text are labelled alphabetically.
Platform Houses Pennant Williams and Leach concentrated their excavations on a group of building platforms (marked ‘A’ on the plan) of a type commonly assumed from excavated examples elsewhere to be medieval in date. This group forms an irregularly spaced row occupying the south-facing slopes below the summit of the plateau at the SE corner of the site, the remains of a possible contemporary track running E-W and immediately below them. Their distinctive earthworks are of a type found throughout Wales (See Diary Entry 1), consisting of an uphill ‘hood’ cut into the natural slope, the material being used to level out a rectilinear, sometimes raised structure-bearing platform, often with a scarped or terraced ‘apron’ or bank defining the downhill end. Their long axes are at 90° to the slope in order to reduce the extent to which the building was exposed to ground or surface water.
House 1 of the 1960s excavations was the most easterly of at least four such buildings arranged in two pairs which are separated by a larger platform or yard. We suspect but need to confirm following ambiguous results this year that House 2 was the next structure to the west. The excavations confirmed surface indications that the Hen Caerwys houses consisted of limestone footings or ‘dwarf walls’ supporting what were assumed to be timber superstructures, the builders having simply utilised the raw materials available to them. The two excavated structures were typical in plan with rounded corners and entrances in the long walls, internal drains in the floor of House 1 seemingly confirming problems with water. This house had a long, narrow annex attached to its SE corner, from which the majority of allegedly almost 1000 potsherds, assorted metal finds and metalworking debris and slag, now mostly missing, are reported to have been recovered. Two aspects of this lost assemblage are remarkable. Firstly its size; this is to our knowledge the largest known to have been recovered from any medieval house in Wales. Furthermore, the finds were apparently almost exclusively of a 14th or 15th century date, the examples illustrated in the surviving excavation indicating an unusual variety considering their humble context.
Towards the SW corner of the site are a second group of platform houses (B). Not only do these occupy an almost identical position to group A on the same south-facing slopes but their overall layout is also similar, comprising two pairs of deeply incised platforms separated by a possible larger structure with its long axis running parallel to the slope. What may be the same E-W track runs immediately below the platforms and there are further terraces and scarps to the W, possibly indicating more buildings. These have not been excavated and are as yet undated.
Fields, boundaries and tracks At least some of the sprawling network of banks and collapsed walls enclosing former fields and hollow ways crossing the site are likely to be contemporary with the two groups of house platforms. At present we can only speculate as to which ones.
C consists of a low, broad and stony bank forming a large rhomboid enclosure, generally corresponding to the route of the main marked trail through the woods. Its interior is much overgrown and disturbed by later quarrying but was clearly subdivided into smaller paddocks or fields. The eastern boundary terminates just to the N of the strongly embanked enclosure H, excavated in 2011. Following the same alignment, an isolated length of bank runs from the southern side of H to the edge of the escarpment; this may suggest that H postdates F but this relationship requires closer examination.
In the large field belonging to Marian Trefedwin, another longer boundary (D) is clearly visible as an irregular stony bank, seemingly of some antiquity. It seems to have formed the south-western edge of a series of small rectangular fields extending to the NE, now mostly surviving as earthworks. A number of other stretches of bank of similar construction (E) have largely disappeared under vegetation since the 1994 survey but apparently formed further small enclosures and paddocks of uncertain date and function.
Also traceable within the woods are series of tracks and pathways. We have noted the tracks that run immediately to the south of both sets of platform houses. The most prominent is a stone lined NW-SE hollow-way (F) running immediately to the east of the enclosure and house site I and is bordered by the collapsed remains of dry stone walls. This leads downhill towards another enclosure and house (J), located adjacent to the present road at the SE end of the site. Another walled track (G) follows the present path on the northern side of the field system C in Coed y Marian and must have become obsolete when the woodland was established.
Enclosures and long houses Whilst many of the more extensive embanked features at Hen Caerwys clearly relate to field systems of various periods, at least two form distinct, more substantially constructed enclosures. Associated with these are the remains of further buildings of markedly different construction to the platform houses.
The large rectangular enclosure (H) investigated in the July 2011 trial excavations lies on and near to the southern edge the plateau. An area of approximately 0.2 hectares is defined by a substantial stony bank up to 6m wide and 0.5m high above the exterior, shown to be a simple dump at the point sampled in our trench but with no signs of an external ditch. The slightly raised interior of the enclosure is almost level and notably featureless, with the exception of a lesser, later bank (L) which crosses it from east to west. The entrance may have been in SE corner where there is a break in the bank. A resistivity survey of the interior of the enclosure in January 2011 located an area of high resistance within the NE corner, which was interpreted as the possible remains of a building. We now know following excavation that the anomaly represented a shallowly buried area of bedrock. Theories abound regarding the date and function of the enclosure, ranging from Roman or Iron Age defensive works to a post medieval stock corral. The issue was not resolved by our 2011 trench and will be hopefully be explored in more detail in a future diary and indeed, season of excavation.
Immediately east of H is a second, smaller enclosure (I), roughly square in plan and containing the remains of a substantial rectangular building with a second elongated N-S aligned structure at right angles to it, forming the western side of the enclosure and possibly linked to the entrance of H by a short bank (here the 1994 plan appears to be incorrect). These structures contrast with the medieval platform houses in that they are not scarped from a slope but are built on the level ground of the plateau, implying a different date or function. The eastern side of the enclosure is bordered by the stone-walled track F. The footings of third elongated rectangular structure (J), also not a platform building survive in the southern corner of Coed Gerddi-Gleision, separated from house group A by the lane to Pant. This appears to be associated with another rectangular embanked enclosure but the whole area has become much overgrown since it was surveyed in 1994.
Later features Whilst the remains described above are difficult to date and relate to each other, the latest features at Hen Caerwys can be identified with relative ease where they sit over or cut earlier ones. A series of shallow and irregular quarries (Q) are scattered across the site, many of them hidden in the undergrowth and providing good cause to stay on the cleared paths when exploring. Some of these clearly post-date the field system as they interrupt the lines of banks and tumbled walls, notably in the centre of field system F. Adjacent to one of these workings In the NW corner of Coed Gerddi-Gleision are the very ruined remains of a probable lime kiln (K), the partially buried arch of its lower chamber surviving. It is difficult to ascertain the age of such a ruined structure but it is likely to be contemporary with the quarry workings and could well repay clearance and excavation with a date. Also late is a slight boundary bank running roughly E-W through Coed Gerddi-Gleision crossing through and probably over enclosures F and H and possibly I. Almost certainly the most recent features of the site are the straight roads which now divide the scheduled area and date from the enclosure of the common.
Summary Whilst we have avoided speculating upon specific dates or functions for the many and varied parts of Hen Caerwys, this quick summary has hopefully raised a series of issues for you to consider. Understanding the chronological relationship between the two groups of platform houses - or indeed, the houses within the two groups - is fundamental to our interpretation of the site. Should they be contemporary, Hen Caerwys may indeed represent a very rare example of a nucleated medieval rural settlement in NE Wales or may simply represent an extended sequence of activity on a favourable parcel of land. Similarly, their relationships to the larger enclosures and longhouse-type structures have similar implications; are we seeing contemporary structures of different function and status or gradual changes in domestic traditions or land use? Can we demonstrate which structures are contemporary with the field systems and roads? At least some of these questions will hopefully be addressed by further fieldwork.
Place- and site names can be fascinating. Some go back many centuries and have obscure origins that can only be guessed at, while others are, or at least appear to be, straightforward.
Compare Buckley, near Mold, and Denbigh for instance. Hywel Wyn Owen and Richard Morgan in their recent Dictionary of the Place-Names of Wales call Buckley a ‘troublesome place-name’, starting with Bocleghe in 1198 and appearing as Bukkelee at the beginning of the 14th century. These varied past forms of the name could have originated very differently. However, they currently have a meaning which when translated from Old English would be ‘the wood of Bocca’, though who Bocca was will probably always remain unclear. Denbigh looks to be much simpler with Dinbych in 1269 and Dunbeig’ in 1211 containing the elements din and bych meaning ‘little fort’ and thought to refer to the motte and bailey castle a mile to the south of the town.
Offa’s Dyke is in a different class. The earliest naming of the great earthwork – as Offedich - comes no earlier than a 13th-century land grant, yet the first link to the 8th-century king of Mercia is much earlier. This is because when Bishop Asser wrote his life of Alfred the Great in the late 9th century he noted in passing that Offa had built a great dyke ‘from sea to sea’. So there is a clear link, but it is based on an assumption.
Buckley, Denbigh and Offa’s Dyke are all major features in the Welsh landscape and it is of no surprise that we have names for them going back into the medieval centuries. For the vast majority of archaeological sites their original names have not survived, and the names that we know them by today are the ones invented by those who discovered and recorded them. Thus if I discover a previously unrecognised medieval farmstead in the hills, I will probably name it after the hill or common on which it was found, or after the stream that passes close to it, or the nearest modern farm lying below the hill. And once the name is entered into the archaeological record, it is generally there for good.
And so to Hen Caerwys. We can assume that this was the name given to it by Ellis Davies or Wilfred Hemp soon after its discovery. Implicit in this name is that it was the predecessor of the ‘new’ town of Caerwys, perhaps even the settlement from which people were moved to establish the new town. Cadw’s records contain an interesting exchange of notes between the landowner, Mr Owens, and Christopher Smith, the Welsh Office Inspector of Ancient Monuments, Mr Owens pointing out the earthworks were more likely to represent the former township of Trefedwin on the evidence of early maps and the surviving name of the neighbouring farm. The Inspectorate conceded that Mr Owens probably had a point but decided that the existing name was sufficiently embedded in the archaeological literature that altering it would only cause confusion.
But the excavations in the 1960s have demonstrated that the name is misleading. The town of Caerwys was in existence in the 13th century, receiving its first town charter from Edward I in 1290. But ‘the men of Kayroys’ were referred to fifty years earlier, in 1242, and this hints at the likelihood that Caerwys was already in existence, a Welsh town later taken over by the English king. As a contrast, the pottery and metalwork from the Hen Caerwys excavations reveal that it was occupied up to two hundred years later.
So Hen Caerwys is a misnomer. Town and settlement existed side by side by the 15th century, and there is absolutely no indication, at least as yet, that the latter came into existence before the former. This, though, is not a plea to alter the name. Hen Caerwys it will always be.
In one of our earlier diary entries Will Davies conducted the reader around the woodlands at Hen Caerwys, providing a brief guide to the earthworks there. In the next couple of diary jottings, I want to look in a little more detail at some of the features that make up the complex.
We introduced the idea of platforms at the beginning of this project diary. Just to re-cap, platforms have long been recognised as one of the fundamental features of medieval settlements in most of Wales. They were first identified in Glamorgan in the 1930s and more examples were found in Caernarvonshire and Merionethshire in the 1950s. Fieldwork in areas of mid-Wales in the 1990s produced lots more examples and revealed a reasonably even distribution across Wales, other than in the lowlands of the south-west, in lowland Glamorgan and Monmouthshire, and in Flintshire and the Vale of Clwyd.
Excavations in the 1930s soon showed that the platforms were used to provide a level stage for buildings. Initially at least, these types of earthwork were known as ‘house platforms’ in the archaeological literature. Sometimes the foundations of the building were visible on the platform, sometimes it was only excavation that revealed them. And increasingly it was realised too that in many places occupied farms and other houses still sat on platforms. Where they appeared simply as earthworks, the fieldworker might find solitary examples or occasionally pairs, more rarely three platforms together. A large platform at Cefn Graeanog in Caernarvonshire was even found to have had four buildings on it when excavated in the 1980s.
The platforms that G. B. Leach and T. T.Pennant-Williams started to excavate at Hen Caerwys in the 1960s showed few surface signs of the buildings that had once occupied them, but when the vegetation was removed and the topsoil was stripped their outlines became rapidly apparent. What was rather interesting was that the first house to be excavated lay at right-angles to the contours and in this respect was very typical of platforms across the country. But they then found that another building had lain at right angles to the first, and so lay along the contour, as the plan in Rogers’ excavation report makes clear. Such a position was altogether more unusual. This second building does not appear to have been a house, but was probably an ancillary building such as a barn or byre, so here we have a small farmstead complex.
Distribution of house platforms at Hen Caerwys
One complex of this kind would be unusual, and I can recall very few comparable examples in the hundreds of platforms that I have encountered in north and mid-Wales. But within a few metres to the west there is another pair of platforms set down the slope with an adjacent one at right-angles (for all of these see the accompanying sketch which shows the locations of the platforms in relation to the modern road through the southern part of the site). And in the second group of platforms 230 metres to the north-west, but in a similar south-facing location there are two more complexes, one certainly and the other probably having buildings set in an L-shaped pattern.
So we appear to have here a rather distinctive type of medieval farmsteading, almost entirely without parallel, at least for the present. Other examples I have no doubt will turn up in due course, but we can, I think, legitimately refer to the Hen Caerwys platform type as a specialised layout, deliberately planned to offer a specific and repetitive form to the farm. Why such a platform layout should be chosen is an intriguing question.
In the last project diary entry we looked in more detail at the two groups of platforms on the south side of Hen Caerwys. Let’s now turn our attention to the enclosure whose bank we put a trench across in the summer. Its rectangular with rounded corners and as anyone who visited the excavations will have seen it’s a pretty distinctive feature, very obvious if you’re walking the footpath that follows its ditch on the uphill side. Its other characteristic is that it appears to be completely devoid of any features in the interior.
Later enclosure at Hen Caerwys
What the general plan of the area shows (see the entry for 28 September) is that the enclosure appears to overlie a boundary bank which forms one element of the complex of enclosures and fields on the plateau top. This boundary bank is a slight yet traceable earthwork to both the north and south of the enclosure but is absent within it – the obvious conclusion is when the enclosure was constructed its builders levelled the boundary bank as it was no longer required. This then provides us with two phases of activity at Hen Caerwys. If the fields and enclosures are associated with the platform houses (and this can be no more than an assumption at present), the rectangular enclosure is from a later phase.
We can then make another assumption. The rectangular enclosure looks to be accompanied by another enclosure immediately to the east, and sitting immediately outside the entrance to the former (see accompanying plan where both enclosures are shown in grey). This is nearly square in outline and contains the grass-covered foundations of a rectangular building (in black), which with its clearly marked compartment walls was evidently a house. But this house is different from the platform houses that we considered in the last diary entry. It’s on reasonably level ground and therefore there was not need for a platform. It’s accompanied by a small almost square structure about ten metres to the north. What the general plan (above) doesn’t reveal is that there is a raised rectangular platform lying at right angles to the house and only a few metres from it. So, then, we have an L-shaped layout to the holding, comparable with those lying just below the plateau edge which were discussed a fortnight ago. But the difference is that complex lies in its own enclosure, implying a solitary farmstead that is not part of a nucleated group.
The final assumption is that this could be another example of how settlements disintegrated later in the medieval period, with nucleated communities being replaced by single farmholdings. It is a concept that has found support from historians and historical geographers over the years, but it has not been possible to point to surviving physical remains which confirm the theory. However, there is one settlement site in western Radnorshire that shows it happening, and Hen Caerwys looks like another.
Woodland Management at Hen Caerwys and Coed Trefraith
That we can explore the remains and woodland at Hen Caerwys is due largely to almost 20 years of carefully considered and often back-breaking management work carried out by Mike Owens, the latest in the line of his family who manage the woods. Here Mike talks us through some of his work, which has benefited not only the archaeology but increased the biodiversity of the woodlands.
During the early part of the 1990s the former Clwyd Archaeology Service and Cadw commissioned an EDM survey of Hen Caerwys and a nearby enclosure in Coed Trefraith, also a scheduled ancient monument in my care, both of which lay almost hidden in neglected woodland and were almost impossible to understand on the ground. I was employed to clear undergrowth and lines of sight for the Gwynedd Archaeological Trust archaeologists to enable them to carry out their surveys. As the work progressed, the lumps and bumps in the woodland were transformed into banks, enclosures and house sites and once I’d ‘got my eye in' I began to spot remains everywhere, the two experts showing a massive amount of patience investigating my new discoveries, which more often than not were natural formations or hollows left by long-rotted root plates. Once the surveys had been completed, Andre Berry (the Clwyd Archaeology Service officer who had initiated the renewed interest in the sites) wrote a management plan for them. This eventually became a Cadw management agreement which is still running. Andre is both an archaeologist and a trained botanist and was once a Coed Cymru officer, providing him with a range of knowledge and experience ideal for tackling wooded monuments.
The woodland at Hen Caerwys had been neglected since the end of the First World War and was very dense and tangled with outgrown Hazel coppice and many multi-stemmed trees which had re-grown from long felled stumps. There were also several open, brambled areas where groups of Elms had succumbed to Dutch Elm disease. Windblown trees were widespread. The scheduled area covers the whole of two large woodland compartments and part of another, so the plan for the site had to have an approach which would be beneficial to both the woodland and the archaeology. Both of the sites presented similar problems. Whilst the woodland had no doubt protected the earthworks from mechanised agriculture, the remains were slowly being destroyed by root disturbance from growing trees and by wind blow, with roots of fallen trees tearing up sections of the banks. It is said that rather like a ploughed field, a wood will turn itself over every 1000 years.
Five areas of Hen Caerwys were identified as being of particular archaeological importance and singled out for intensive management. These were the two large enclosures H & E, also described by Bob in the last post, both sets of house sites (A & B) and the large encircling bank of the field system (C) (see the plan and descriptions in the 28th September and 10th October diaries to remind yourselves where these features are). These were subjected to an annual strimming regime. The effect of this was to stop any further regeneration of trees and to keep the key areas accessible. The eventual aim was to create grassy glades within the woodland but this didn’t happen as the bramble and Rosebay Willow herb quite enjoyed their annual pruning and kept springing back.
The larger scale management was carried out in stages over fifteen years. Firstly, old, neglected Hazel stools, many of which were dying off or being blown over, were coppiced. I remember cutting down the ancient coppice stools in the large enclosure, some of which were over four feet across and produced massive amounts of timber and brash which needed to be shifted off site. At the time I hadn’t yet invested in a machine so I used an old trotting sulky, loading it with cut pieces and taking the place of the horse myself. Each time I made it to the edge of the enclosure I would collapse in an out-of-breath heap. Needless to say it wasn't long before I bought a Crawler Tractor. This gave me plenty of power to extract any tree I needed but with the advantage that it would not sink in and cause a rut even in the wettest conditions. Most of the rest of the wood was then thinned and I felled all of the poor trees, coppiced the under-storey and "singled up" the multi-stemmed trees. The roots of these, used to holding up three or four large stems, become much less prone to wind blow when supporting just a single stem. Since this task was completed Hen Caerwys has hardly ever suffered from wind blow. The five intensively managed areas were also thinned but more intensively, leaving them as open woodland leaving only the strongest trees and making it possible to see the earthworks.
Thanks to Mary Erasmus for this photograph of an orchid in the Hen Caerwys woodlands.
The next step was public access. As part of a joint project funded by Cadw and Cadwyn Clwyd, a circular path was created around the woods taking in the major remains on along its route. A picnic area was created in woodland just outside the scheduled area and the public were actively encouraged to visit the site with the production of a leaflet and a notice board based on the 1990s survey.
Four years ago we replaced the annual strimming with a cut in the winter and another in late summer, keeping the site accessible for a larger part of the year. A surprising result of this was an increase in Flora. Hundreds of species of wild flower now cram the open areas in the summer and it makes me feel terrible to decimate this beautiful site every July. It is, however, the act of cutting which encourages this diversity, these areas being akin to an old-fashioned hay meadow or even woodland pasture. This echoes a process which would have taken place in the past due to animals grazing within the wood, something which perhaps wrongly is no longer encouraged today.
As for the future? Whilst the aim of the work is to maintain the archaeology as it is, something which is now achievable, the woodland will continue to change and the canopy is already closing up again on all but the five main areas. This shows us how the management of woodland archaeological sites is an ongoing task and it would not take long for the Hen Caerwys to revert to its pre-1990s state if work stopped now. The coppice is mostly in rotation (all I need is a decent buyer or someone who wants to work it) and visitors are coming to see the wealth of biodiversity as well as the settlement. I would be happy to extend the intensive management to the rest of the remains, which would not only guarantee the future of site in its entirety but also create a series of wildlife corridors throughout the wood.
Janet Smith is a geologist and also chairman of the Dyserth Field Club. She visited Hen Caerwys during the excavations last summer and we asked her to put together a few words for the diary on her observations.
The revelation of a pristine limestone pavement at Hên Caerwys is not only exciting for archaeologists, it is also an important geological find. While it was known that this part of Flintshire was underlain by limestone, the form of that surface could only be assumed – if, indeed, it was thought about at all! The following is a broad outline of its origins.
Limestones, and there are many types, have been called ‘chemical sediments’ because, whatever their formation, their content is ‘lime’, calcium carbonate in various forms. It is because of this that limestones react chemically with rainwater since the rock is alkaline and the water acidic and so the rock is slippery to walk on after rain because surface solution will have occurred. Over time, and especially when temperatures are warm, that solution will work on the limestone beds/layers to deepen and widen the cracks, or grykes/grikes, between the blocks or clints of limestone. At the same time solution will also work on the surfaces of the clints, deepening furrows and leaving the ridges between them interestingly shaped! The overall effect is to increase the entire surface area of the limestone. Ultimately this process erodes deeply into the rock and, as long as the beds are not steeply tilted, subterranean erosion will produce cave systems, also of interest to archaeologists!
These landforms are aspects of limestone scenery which tends to be a very dry environment when it is fully developed because surface water will have disappeared underground. It is worth noting here that such areas are also of great interest to botanists since plants are confined to the grykes which are cool and shaded areas where moisture will gather and soil will eventually develop, providing micro-climatic conditions for plant growth. The name karst is used for this scenery by geomorphologists*. It comes from The Karst, a limestone plateau in Slovenia at the northern end of the Dinaric Alps. The development of this type of landscape in that area is some of the best of its type in the world.
*those who study landscape i.e. what used to be called physical geography
The saga of Hen Caerwys and the early excavations there is one that we can only put together, piece by piece. Debbie Seymour of Flintshire Museums Service takes up the story..
A few days before I was due to get married, in July 2008, I received a phone call; a Mr Leach was clearing out an elderly relative’s house and wondered whether the Museums Service wanted the items which seemed to have belonged to G. B. Leach, the Flintshire archaeologist? He was going back to South Africa in a few days and anything we did not want would go on the tip. I knew very little about G. B. Leach but enough to know that I was interested, particularly in the ‘bits of pottery’. I arranged for Mr Leach to leave anything of historic interest on condition that I could dispose of anything we did not want.
When I returned from honeymoon, I found my office full of old suitcases and boxes and I began to regret my initial enthusiasm! It mostly consisted of the library of George Leach, but there were a few boxes of interesting looking bits of pottery some labelled ‘Caerwys’. Over the next year or so (including 6 months maternity leave), I gradually listed the piles of ‘stuff’.
It turned out there were 64 journals, 75 books and 40 offprinted articles, a quantity of archival material including a plan of George Leach’s garden and a small number of excavated objects. Much of this material was not relevant to Flintshire Museums Service and I began the task of thinking who else might want it. I transferred some Flintshire related pamphlets and books to the Record Office, and two fascinating autograph books from a Mabel Earpe’s time at the Military Hospital in Frodsham in World War I to Cheshire Military Museum. Some of the pottery marked ‘Chester’ and ‘Professor Newstead’ I gave to the Grosvenor Museum, Chester as well as some of the offprints signed by Prof Newstead. I gave all the old suitcases to a local amateur dramatic society!
I was left with a boxful of published articles by G. B. Leach, miscellaneous sherds in a box marked ‘Caerwys Pottery 1968’ and annotated with cryptic location details such as 'On floor near Bloomery 1 1'1" below surface'., and an unfinished typescript article about Caerwys together with related excavation ‘plans’. The minimal research I did established that Leach had excavated at Caerwys with Pennant-Williams in the 1960s, but that nothing had been published apart from the much later article by T. Rogers. I was surprised that, although George Leach seemed to have been very active archaeologically, there was very little information in the Museum Service collection or at the Record Office. We had some finds from Hen Blas, but nothing else. The Caerwys sherds hardly constituted a site archive and I wondered where the rest had gone. I decided to accession and catalogue the material we had, and hope to shed some light on its significance and history in the future. So, I assigned the group of items the accession number FLIMS 2011.4 and catalogued 45 separate items or groups of items on our database. I put everything, labelled neatly with its number in a large box and took it to the museum store.
It was shelved alongside other archaeological archives of other Flintshire amateur archaeologists from the 1930s to 60s such as Gilbert Smith, who discovered Prestatyn bathhouse, and William Stead who excavated at Gop Cave. These archives are mostly partial and poorly documented, but I am fascinated by the work these amateur enthusiasts did and the glimpse these partial archives give of their world is tantalising – signed offprints circulated amongst authors, notes about supplies bought for excavation work, scribbles in the margins of books and finds stuffed into old cigarette packets and wrapped in old newspaper.
So, I was delighted when Will Davies from Cadw enquired about whether we had anything from Caerwys and I could actually say yes we did, although it was fragmentary and frustrating in its lack of context. But even this could have so easily ended up in a skip.
Hopefully, through this project, some more light will be shed on the history of Caerwys and these fragments of pottery dug up in 1968 and stuffed into brown paper envelopes will be dated and drawn and assessed as they should have been in 1968.
Debbie Seymour Principal Museums Officer
Flintshire Museums Service
We’re not going to post a text on Hen Caerwys this week, largely because both Will and I have been too heavily involved with other archaeological sites in north-east and east Wales to have had a chance to sit down and think about HC, much as we might like to.
However, what we can report is that we are now starting to plan a further ‘season’ of work at HC in 2012.
In 2011, we were on site for five days during July. This year we hope to expand that to ten days spread over a two-week period which will coincide with the Festival of British Archaeology which runs from the 14th July to the 29th July. Its early days as yet, but provisionally we hope to finish off the two excavations that we started in 2011, by completing the section through the bank of the big enclosure, and excavating the lower end of house 2. We also hope to open a couple of new trenches on parts of the site where there are questions to be answered, but the scale of this work will depend on the number of volunteers and also getting all the right permissions to work on the site. And again we hope to be open to visitors throughout the time that we are there.
So if you are interested in joining us during July, for a day, a few days or even the whole time, do get in touch.
Reporting Hen Caerwys:
Part 1: The written record and setting it straight
In our 19th January post, Debbie Seymour provided us with a colourful account of the fortuitous journey of the remains of the original 1960s’ project archive into her care at Flintshire Museums Service. This week we will take a closer look at what remained of the notes and drawings of the excavators and how some of these would resurface some years later in the only illustrated published report on their work.
When Leach, Pennant-Williams and their fellow members of the Flintshire Historical Society began excavating the first of a pair of house platforms at Hen Caerwys in the early 1960s, Pennant-Williams spoke to a local newspaper, outlining ambitious long-term plans for what was effectively a week-end project, stating that:
‘Work on this house... ...should be completed in four months’ time. Then we will cover it up again to preserve it and start excavating another’ presumably referring to the structures we know from the surviving records as Houses 1 and 2. He continued: ‘And so it will go on until the whole village has been unearthed. It’s a big job and will take a number of years’
This would indeed have been a ‘big job’, requiring not only many years of excavation and careful recording (given that two houses took the team at least five years to uncover) but also a comparable amount of time processing and analysing finds, producing a site archive of drawings and other illustrations and ultimately a length report, supported by the relevant background research.
It was perhaps inevitable that such ambitions were never realised and the weekend visits petered out by the end of the 1960s. We are uncertain exactly why this was the case; unless there are further records of which we know nothing, work seems to have ceased at least five years before the deaths of Pennant-Williams in 1973 and Leach in 1975. The immediate legacy of their work, a series of un-illustrated interim notes in the fledgling Archaeology in Wales journal, may well have disappointed both archaeologists. Had the excavations been completed to even half of the intended extent and then published, the true importance of Hen Caerwys would undoubtedly have been recognised at an earlier date and would still stand as one of the few extensive studies of a deserted medieval rural settlement in north Wales.
What we do have are tantalising fragments of what was evidently a much more extensive archive of material. The collection identified by Debbie Seymour as belonging to the late George Leach comprises set of typed and handwritten notes relating to House 1 and its annex, a handful of pottery sherds, a series of detailed site plans, drawings and accompanying descriptions of individual finds, some frustratingly incomplete and anonymous historical notes (possibly by J.E. Mesham) and a number of miscellaneous items. The latter include a small notebook detailing project expenses down to the cost of cucumber sandwiches and an almost illegible draft of a letter from one of the excavators to the archaeologist John Hurst, who along with his historian colleague Maurice Beresford was a pioneer of the study of deserted medieval settlements in Britain. The duo had been directing the famous excavations at Wharram Percy in Yorkshire since the 1950s, their collaboration culminating in the standard textbook ‘Deserted Medieval Villages’ and were obvious experts for the Hen Caerwys excavators to consult. Whilst such details are incidental to our understanding of the site Debbie has pointed out how they offer a glimpse into the world of what was effectively the final generation of gentlemen antiquarians.
The plans and notes are the most useful items to us. Some are in pencil, dirty, frayed, much -scribbled-upon in different hands and clearly working plans that were used on site whilst others are inked and heavily annotated fair copies showing the excavation at various stages of completion. Final drawings were meticulously drafted by Francon Lloyd (?) Criba (corrections by readers are welcome), a Rhyl based architect, also seemingly responsible for a plan of Leachs’ garden. Most of these are plans and sections of House 1 and to a lesser extent its annex, all published in 1979 by Tom Rogers. More exciting were two working drawings of the suspected but otherwise undocumented partial excavation of House 2, the assumed platform of which we began to excavate last year in order to confirm its identity. This epitomises the frustratingly incomplete record available to us – we have no accompanying notes for this building, nor any overall site plan to help us to locate it on the ground. Nor can we tell whether the few available finds belong to this house, House 1 or another structure again.
The more extensive typewritten notes, some heavily annotated clearly formed a report in preparation and deal exclusively with the excavation of House 1 and its annex. Another copy of this document, this time in the Pennant-Williams collection at the Flintshire Record Office, demonstrates that whoever was responsible for the bulk of the text (this is unclear), the two evidently passed drafts between them for editing. We have also seen earlier, hand-written drafts of the same work. It is also notable that the site drawings also appear to have been gently modified at various stages.
The typescript would eventually come to light in 1979 when the excavation of House 1 and its annex was published as a short article by Tom Rogers in the Bulletin for the Board of Celtic Studies 8see below for a reference). This was apparently based upon the site records and available finds, Rogers noting that the ‘great majority’ of the pottery was ‘not available for examination’ and had clearly been dispersed by this point. We do not know whether the surviving papers and finds in the care of Flintshire Museums equate exactly with those seen by Rogers. The loss of the most of the finds is very unfortunate given the unusual volume of material and variety of metalwork and stone objects illustrated by Leach, although his notes and drawings of these were clear enough for Rogers to recycle in his report. Read alone, the report itself is an unusual piece. Aside from well-reproduced and tidied up versions of Leach and Pennant Williams’ plans and sections of the house and annex, it lacks any description or illustration of the rest of the site, effectively being a dry commentary on the excavators’ finds without any attempt at interpret them or place them into context. This was noted during our own research prior to last years’ dig and the discovery of the versions of the original typed manuscripts, our assumption being that Rogers had probably salvaged a mess of since lost material and made it publically accessible in a short report.
An example of part of the report archive
It would appear that this was far from the case. At the close of his article, Rogers acknowledges the assistance of Leachs’ daughter and Pennant-Williams in providing access to the finds and co-operating in the compilation of the report. A more correct gesture would have been to accredit Leach and Pennant-Williams as co-authors of the article, which although not a word-for-word copy, rarely deviates from the original typescript draft of the excavators. This is in effect their work, as are the plans and most of the finds illustrations. We know this having seen the originals, firstly annotated by hand then typed up; like the report text, they have simply been re-arranged. There are, however, a number of small but significant details that were overlooked
Apart from having published part of the results we are not entirely certain of Rogers’ connection to the original excavations, although several reliable sources have informed us that he may have conducted some work somewhere on the site in the 1970s with or without the Owens’ permission. As we have mentioned in an earlier post, he certainly had bigger plans for Hen Caerwys, as detailed in a series of letters to Cadw’s Welsh Office predecessors who gently deflected his proposals to excavate it, after which his interest waned. Regardless of the ethics of Rogers’ publication of the work as his own, it is fair to say that its appearance in 1979 helped to prevent this wonderful monument and the careful work of our gentlemen archaeologists from disappearing into obscurity, as has been the case for so many other unpublished excavations.
Next week we will take a closer look at Leach and Pennant-Williams’ records, Rogers’ publication and the information that we can extract from them. Again, our contact details are posted below if you have any comments or wish to participate in the 2012 excavation.
As discussed in our previous diary entry, Rogers’ 1979 article on the excavations at Hen Caerwys is essentially a re-wording of Leach and Pennant-Williams’ original notes, although he missed or ignored several important pieces of information. This week we will begin to piece together the facts and briefly explore the first of the excavated buildings.
The first thing that struck us when reading Rogers’ report was the almost complete lack of a pre-amble or introduction to the site beyond the story behind its initial identification and the fact that it was excavated by Leach and Pennant-Williams in the 1960s. He provides no location map, plan, description of the site, nor any historical and archaeological background, but simply launches into an account of House 1. Adding almost nothing of his own observations or original research, this entirely reflects the material at Rogers’ disposal although the ambitious plans outlined by Pennant-Williams to the local press suggest that that further work must have been intended at a later stage. That Rogers had applied to several institutions for funding to carry out a site survey at the end of the 1970s indicates that this is not just a case of the work being lost. Similarly, we have referred to several pages of detailed historical notes which were clearly to be expanded upon.
Excavation plan of House 1 and its annex. Rogers after leach and Pennant Williams.
House 1 This deeply incised, partially rock-cut platform lies at the SE extremity of the site, its sharp contours and sunken, level interior being clear signs of an excavation. A low earthen platform extends to the E from its SE corner, marking the position of a large annex. Rogers and the excavators describe the house as measuring approximately 12 x 3.2m. The platform on which it stood was dug and quarried from the slope, loose material being cast downhill to form a level base for the building. Some of the naturally occurring limestone was then used to construct clay-bonded stone footings or ‘dwarf walls’, approximately 0.6m thick, which survived to a height of a few courses. These have since partially collapsed but can still be seen in places. Probable entrances were located at the southern ends of the E wall (where two posts define a gap) and possibly opposite in the W, a surface of crude limestone pitching being mentioned but only vaguely recorded outside these.
Within the house a central line of stone-lined post-holes will have held posts carrying a roof. It was suggested that this was thatched in the absence of any evidence for tiles or turves, although it is often difficult to identify the latter archaeologically. At a later stage, a platform with a clay floor was constructed at the uphill end of the building. This was supported by a low wall, possibly marking a partition dividing it off as a separate room. As might be expected for a building set in a deep cutting at the base of a porous limestone slope, water was evidently a problem, whether percolating through the ground or running off the surface. Evidence of this is provided in the medieval solution, a drain and sump being dug into the clay floor of the possible upper room.
The Annex A rectangular stone ‘annex’ measuring roughly 13m long x 3m wide and divided into two rooms was added at a right angle to the SE corner of the house. In spite of the fact that by far the largest assemblage of late medieval finds to be recovered from any Welsh house site to date came from this building, Rogers’ description is for some reason restricted to a few sentences, the surviving excavation notes, whilst also very brief, providing us with much more detail. That it was an addition is demonstrated by the fact that the annex walls were apparently not bonded to the house, although we have no idea how much later it may have been added. It rested on the same layer of limestone chippings as the lower end of the house, this time containing cockle shells, implying that it was founded on earlier domestic debris. The excavators again assumed that the walls supported a timber superstructure due to the lack of surface rubble. No internal posts were recorded.
Sited parallel with the slope, the annex predictably seems to have been affected by ground water, a large drain (absent from Rogers’ report) being dug into the excavators’ ‘Room 1’ to the W and its presumably soggy floor covered by small pieces of limestone. Above this, a dark occupation layer of black soil (also ignored by Rogers) had accumulated to a depth of almost 0.3m, from which the great majority of the since lost finds were recovered, including several hundred pieces of pottery, metalwork and quantities of coal and iron slag, shown by ‘analysis’ to come from a coal forge and leading the excavators to suggest this use for the room. The eastern ‘Room 2’ evidently had a different function and was largely devoid of occupation evidence, just four potsherds and some metal objects being recovered from its floor. Rogers also failed to describe two large openings in the southern wall of Room 1 (and possibly one opposite) and another in the northern wall of Room 2. A post in the middle of the Room 1 opening (Post 12 on the plan) was assumed by the excavators to have supported a wall plate for the roof over the gap in the wall. Such a large opening may support their interpretation of an industrial use for this room. It is unclear from the notes whether there was direct access into the main house – no walling is marked on their plan but a wooden partition may have existed.
Earlier beginnings? Lacking the finds themselves we have to assume that Leach and Pennant-Williams’ notes and conclusions were generally correct and that the majority of the objects recovered and therefore the buildings dated from the fourteenth and fifteenth centuries. Whilst the notes for a number of items speculate on earlier, even Roman dates this cannot be confirmed but the excavators provide us with several clues that House 1 was more complex and possibly older than the simplified published plan suggests.
The excavators state that the southern end of the house rested on a layer of limestone pitching, presumably similar to that outside the walls although they are not clear whether they thought this belonged to an earlier structure. It is notable that the annex to the E also stood upon this layer, where the notes record that it contained cockle shells which were presumably debris from earlier activity on the site. Internally, there was more convincing evidence, the heat-cracked remains of a hearth consisting of a slab of sandstone with a blackened fragment of a limestone surround being excavated beneath the floor of the building near the sump. This lay directly underneath one of the central line of postholes (Post 4 on plan), although the interpretation in the notes is suspect:
Evidently the hearth was a temporary one, used only while building the house and discarded as soon as post no. 4 which rested on it was erected.
A more likely explanation is that there was an earlier structure on the site or that the central posts belong to a later phase of construction. Another nugget overlooked by Rogers would seem to confirm this:
There was a shift in the roof at the south end. This is shown by Post Hole No. 7 being out of alignment with the other central post holes and had been dug through the occupation layer.
These features could indicate one of two things. Either that the whole structure had a predecessor or that the posts were related to a later refurbishment or alteration of the house. We cannot be certain which interpretation is correct but it is clear that the structure excavated in the 1960s had a more complicated history than Rogers recognised and that future excavations on the site may confirm this.
Next week we will look at the evidence for, the second house.
Continuing from our last rather long entry, this week we will take a look at the slender surviving evidence for the rest of the 1960s work.
House 2 The short notes published in the 1962 (where Bevan-Evans appears as Pennant-Williams’ co-author, to be replaced by Leach the following year), 1963, 1965 and 1966 editions of the newly founded Archaeology in Wales Journal tell us little that cannot be found in the excavation notes, Rogers’ report and the Cadw archive, the notable exception being the paragraph from 1966.
This describes the start of works on a second structure divided into two rooms ‘close to and south-west of the platform house already excavated’. We initially assumed that the plan of this house survived as a pair of unfinished drawings uncovered amongst the remnants of the site archive and tentatively named House 2. It is curious that Rogers failed to mention the house in his publication of the excavators’ notes when he was almost certainly aware of the earlier work, his correspondence with Cadw’s Welsh Office predecessors outlining his gently rebuffed plan to complete the excavation of a House 2 before moving on to other structures on the site.
Locating this building today is slightly more problematic due to the lack of an overall site plan in the excavators’ notes although the 1966 description and several reliable sources who remember the original excavations indicate the next platform to the SW of House 1. This has clearly been disturbed and probable spoil heaps lie to either side, our 2011 trench being positioned to confirm that the remains of its walls matched those depicted in the surviving drawings. These seemingly indicate a structure of similar form to House 1, with dwarf-walls, internal posts and a possible south-eastern entrance. This building is apparently distinct from House 1 in being a markedly more regular rectangle in plan and somewhat simpler in its internal features, assuming that they were ever fully explored. There are, however several features which cast doubt on the drawings as being of a separate house. One of them is dated to 1963, when we know from the interim reports that Leach and Pennant Williams were working on House I. In addition, a number of key features such as tree roots, a drain by the W wall, the SE entrances and some pencilled-in walls to the E of these share comparable positions to broadly similar features in House 1. It is therefore possible that the plans are early versions of the work on House 1, perhaps reflecting the excavators’ less than objective interpretation of its walls as they exposed their upper surfaces, sketching the beginnings of the annex to the SE. That the available fragments of the site archive contain no other record of the work on the second house might support this theory.
Which House? The as yet unidentified plan.
Unfortunately, our 2011 trial excavation did not fully resolve this problem, although the complete lack of occupation material would appear to indicate and some disturbance. Otherwise no notes survive for Leach and Pennant-Williams’ excavation of this house beyond their two sentences in Archaeology in Wales, which also mention ‘surface signs that there is another room to the south-west side contemporary with the platform house as shown by the pottery’. On the ground, the bases of the walls still stand a few courses high in places, the tumbled lower ‘terrace or apron’ end appearing slightly wider that the platform itself and perhaps accounting for this possible room. Confusingly, the dimensions of 44ft x 38ft (13.4m x 11.6m) given in the interim are wildly at odds with those of the structure visible on the ground, which is a rectangle of similar proportions to House 1, although this may be a typing error. More careful work is planned to clarify this situation in 2012.
Other trenches? The opening paragraph of what would become the published report correctly identifies the complex of earthworks around the excavated house as those of other buildings, briefly noting that:
‘A preliminary exploration on the nearest one (bank) to the house, proved this as foundations and pottery sherds similar to those of the platform house were found’.
The ‘house’ is presumably House 1 rather than the adjacent occupied farm although we have no way of knowing whether the bank in question was part of the House 1 Annex, a reference to early work at House 2 (which would explain the date on the drawings) or another structure.
More worryingly there is some evidence that work may have continued longer than the interim reports (which end in 1966) suggest. The surviving box of finds held by Flintshire Museum Service is dated 1968 (more of these in a later post) and several reliable sources have suggested to us that as late as the 1970s, other, unspecified individuals carried out excavations either in the corner of the field to the S of the western group of platforms or in the corner of Coed Gerddi-Gleision to the W of this group. Whilst firmly in the realm of hearsay, we have included these possible events to illustrate another dimension to the problems the Hen Caerwys poses to us when revisiting it over four decades after the original excavations.
Next time, we will consider the value of Leach and Pennant Williams’ work to our project and what it tells us about their skills as archaeologists. As ever, feel free to contact myself or Bob if you have any points or queries, especially if you can point us towards any of the missing records or finds.
To our readers. Our apologies for the delay in putting this new section on our web site. Other tasks have meant that Hen Caerwys has had to take a back seat in recent weeks. Here we offer the first part in a two-part assessment by Will Davies of the early excavations at Hen Caerwys.
Archaeologists or Antiquaries?
We have devoted large chunks of our previous diary entries to the excavations carried out by Pennant-Williams, Leach and their Flintshire Historical Society Colleagues, the surviving evidence and the problems it presents us. Over the last six months of ad hoc research our ideas have slowly progressed, not only in our thoughts on the site itself but also of its excavators.
Two main issues present themselves when attempting to assess the quality of the work done by Leach and Pennant-Williams. Firstly, it is easy to judge harshly earlier work in any field as poorly executed by comparison with the higher standards that we seek to employ today, but we have to take into account the state of archaeological practice at the time. More fundamentally, it is impossible to arrive at any firm conclusions from the available fragments of what we can only assume to have been a more extensive site archive (i.e. plans, record sheets, field notes and photographs), although the present evidence for their former existence is by no means conclusive.
As the Cadw inspector responsible for the management and conservation of Scheduled Ancient Monuments in the region, I would not be able to support a project with the indefinite timescale and potential outcome of the 1960s Hen Caerwys excavations. Today, under the Scheduled Monument Consent (SMC) procedure, all research excavations on scheduled sites are subjected to rigorous scrutiny in order to ensure that projects are adequately designed, resourced and staffed with realistic timetables and proper provision made for post-excavation work and reporting. As Bob has quite rightly pointed out in one of our earlier posts, many aspects of the 1960s excavations occurred in the wrong order i.e. intrusive excavation occurring before a proper survey had taken place or simply not completed to modern standards.
We must remember, however, that Hen Caerwys was not a scheduled site when the Flintshire Historical Society began digging. Nor was the current requirement of obtaining formal Scheduled Monument Consent for excavation on a scheduled site in place, especially where well-connected local archaeologists like Pennant-Williams were concerned. Whilst their methods may not have matched modern professional standards, the excavators at Hen Caerwys seem to have been methodical and careful, certainly no less so than the academic practitioners responsible for some of the landmark excavations of their time, many of which produced records that no longer stand up to scrutiny.
The surviving drawings and well-written notes reveal an eye for detail, seemingly careful recording, broad archaeological knowledge and a good technological grasp of the structures and finds that they were dealing with, as might be expected of experienced local archaeologists who were Fellows of the Society of Antiquities. The notes accompanying the drawings of the finds indicate the beginnings of detailed research into their origins and backgrounds, Pennant-Williams and Leach drawing on their experience excavating elsewhere in the region and their extensive archaeological contacts. We used one of these drawings as an illustration in our last diary entry, this one containing an intriguing note of April 1967, referring to the spectrograph analysis of the cauldron fragment that it depicts. Whilst no related report survives, this specialist work is also alluded to in Rogers’ later publication and tells us that the excavators were aware of and willing to try newly emerging scientific methods and had enough credibility to approach the relevant institutions to commission it. Any images of rank amateurs are further dispelled by the meticulous surviving account book for the project which indicates that (?) Leach ran a carefully planned and managed operation.
We are not entirely certain of on-site conditions and procedures –the plans document the progress of a systematic area excavation of the interior of House 1, seemingly with sections cut to examine sub-floor deposits. However, the true extent vertical and horizontal extent of the excavations is not entirely clear from the drawings, nor is it specified in any surviving record. At some point in the future we may go back to test this ourselves. Whilst it is likely that at least some original records have been lost, we can be very certain that no full survey of the earthworks was completed prior to excavation. Indeed, Rogers’ proposal to carry out this work indicates that he was not aware of one, in spite of his contacts with the excavators. Whilst almost certainly intended at some stage, the absence of a survey is a fundamental shortcoming alluded to in several previous diary entries and contrasts with a conscientious commitment to the publication of yearly interim notes in Archaeology in Wales. These chart their gradual progress from little more than vegetation clearance in 1962 through to the opening of a new trench over House 2 the following year.
This commitment does not seem to have extended to ensuring that the work was fully written up and published, if only by an appropriate third party, a fault also common to many of their more celebrated archaeological contemporaries. At a recent lecture in Northop to the Flintshire Historical Society, Bob compared Hen Caerwys with a site in Radnorshire known as Beili Bedw near St Harmon which is also renowned for its platforms. It was excavated in 1961-2 by a professional archaeologist but it was not published for another thirty years, and then by someone else who had had no involvement in the original excavations.
On the other hand, in the newspaper quotation from last week, Pennant Williams mentions backfilling his trenches to protect the remains after excavation, an ethos lacking in some much more recent archaeologists. In practice then, in the period shortly before professionalised and fully regulated archaeology with published standards and guidelines, our two excavators fall somewhere within the eclectic range of fieldworkers active in the country, certainly not amongst the best but far from the worst. In short, they were aware of the need for a systematic approach to their work and of recent technical and academic developments but displayed a rather blasé approach to other aspects of their work, and this unfortunately has clouded our present understanding of their site.
What follows is the concluding section on Will Davies' musings on the earlier excavations and reporting at Hen Caerwys.
The great lost site archive?
Before I draw a line under the subject we should also consider the lack of a more extensive set of working records and notes. It is notable that some of the papers retained by Flintshire Museums are copies, implying that the excavators had a set of notes each, the missing one potentially being more comprehensive if we assume these records ever to have been made. When he published those available to him, Rogers implied that there were more records elsewhere, to which he did not have access, his reproduction of the typescript we have seen bearing this out. What exactly happened to them, or indeed to the finds seen by Rogers (again, more were implied), we may never know. However, correspondence between the County Archivist on behalf of Leaches’ daughter and Rogers indicates that the latter was stalling on returning those finds that he had obtained in order to write ‘his’ article. This is odd given the fact that most, if not all of the information that he published on these was lifted directly from the excavators’ notes. The rightful owners were, of course the Owens family.
Whilst we have so far failed to track down the finds described and reported by Rogers, we know of their existence and general form from his publication and the original excavation notes. We can only assume that equivalent material was retrieved from House 2 and recorded in a similar manner although the small box of pottery surviving in the Flintshire Museums Archive is dated 1968 and does not necessarily come from the same structure. Furthermore, as explained in our previous post, we are no longer certain that we even have a plan of this building. The very brief writing up and reporting of the House 1 Annex perhaps provides an alternative scenario to our suspected lost records. One of the later structures to be excavated, the paragraph devoted to the annex in the notes and summarised in a couple of sentences by Rogers is remarkably short given that it contained one of the largest and richest assemblages of finds to come from any medieval Welsh house site. Were standards slipping as interest waned and other commitments distracted the excavation party? The sheer bulk and range of material recovered will certainly have presented a daunting task. Or has a decent site archive simply been lost? We would be happy to have either theory – or others - confirmed and again throw the question to our readers – does anybody remember or know anything about the original excavations and the possible whereabouts of the finds and records? Even the slightest clue may be of use to us. Our contact details are below
For now, this entry completes our discussion of the 1960s excavations, leaving only a description of the surviving pottery stored by Flintshire Museums for next time, once the CPAT experts have taken a proper look at it. Our following entries will move towards future work at the site, namely our 2012 season, details of which will be announced in a separate post next week.
In the previous entry in the diary we lamented the fact that the finds from the earlier excavations at Hen Caerwys in the earlier 1960s and which Tom Rogers published in 1979 have now disappeared. We have asked everybody with an interest in Hen Caerwys about the whereabouts of the finds, but without success, and I think both Will and I are pessimistic about the chances of them now coming to light.
But as Debbie Seymour’s contribution to the diary on 9 January points out there is a small amount of pottery, from the excavations in 1968, still extant and currently housed in the Flintshire Museums Store. From the writing on the envelopes in which it was stored, the pottery was recovered during work in the later part of 1968 (primarily October).
The material, in its unwashed state, consists of 29 sherds of pottery, 1 fragment of glass, and a piece of shell. The annotations on the envelopes provide a little information on context, referring to an occupation level, black soil just outside the south wall, cobbles, and a bloomery. All this sounds as though the material could have been derived from the ancillary building associated with House 1, yet by 1968 we had assumed that the excavators had finished with this platform and moved on to House 2.
The small collection of pottery are mainly fragments from the bodies of jars and jugs, several of them having glaze. Surprisingly, there are virtually no cooking pot fragments. A glazed spout comes from a drinking vessel known as a costrel. Colleagues at the Clwyd-Powys Archaeological Trust have examined the pottery and identified two types of fabric, though their identifications will need to be confirmed by a pottery specialist in due course.
There are sherds of what we term ‘Brenig E’ ware (named from an excavated site on the Denbigh Moors) which are comparable with material collected during fieldwalking on a kiln site at Ewloe, near Hawarden. From other settlement contexts this type of pottery could date from the 14th through to the 16th century. There are also some sherds that could be ‘Brenig C’ ware, which in the Brenig report was dated to the 16th century through comparisons with material from Norton Priory in Cheshire. The comparison with Brenig C must, however, be treated with caution.
The glass shard is from a thin-walled vessel, but little more can be said about it.
It is not a particularly informative collection of pottery, but for the moment it is all that we have got. Hopefully, our excavations in July will provide us with more material.
Sadly, we have nothing much to report at the moment. I’ve been involved in other projects and Will Davies is busy putting some groundwork into the planning of the excavations at Hen Caerwys in the second half of July. But we hope to prepare something in a fortnight’s time.
As many of you may already now, Cadw and CPAT will be digging at Hen Caerwys again this July, details of which will be provided below. Over the next couple of weeks we will give you some insight into the months of preparations and hard work and the decisions that will have been made by the time we arrive on site.
Grand plans and preparations The primary objectives of the 2011 trial excavation were to test both the condition of the archaeology and the feasibility of a community based project at Hen Caerwys. The positive results on both counts were enough to convince both of our organisations to support a second season, the project now being part of CPAT’s main Cadw funded programme.
This project diary has charted our developing ideas on the archaeology of the site as our research has progressed, certain areas beginning to emerge as contenders for informative trial trenching. Planning began in earnest on a bitter day in February when myself and Fiona from Cadw, Bob and Ian from the Trust and Mike Owens met to talk through these ideas on site whilst the vegetation cover over the earthworks was at its lowest. A lot of different things were discussed from the archaeological to the purely logistical, the advantage of a group meeting being that flawed plans and oversights are usually picked up quickly. Our starting point was the two 2011 trenches, which will need to be completed this year and some freezing discussions of exactly where to locate our 2012 trenches. Once an uneven consensus had been reached on excavation areas i.e. one trench was micro-sited whilst others were only generally decided, the conversation turned to practical matters. Having learned from the successes and shortcomings of last year, this conversation covered everything from parking to potential numbers of volunteers and visitors, from health and safety to on-site information boards and from shelters to signposts, all of which need to be organised..
The months since have been spent in a flurry of e mails and phone calls, the final design of the project now taking shape and tasks being provisionally divided up. Extra staff have been commandeered from both of our organisations, the excellent Sophie Watson from CPAT’s development control branch and Caroline Pudney, Cadw’s new CBA sponsored Community Archaeologist. We are very pleased to have them on board to work alongside our volunteers as they are both experienced excavators who have also taken part in a number of community projects elsewhere in Wales. Meanwhile, myself and Bob have been on site with a total station, surveying the two main excavation areas in greater detail than we currently have on plan, in order to make a record before the ground is disturbed and to provide a plan on which to base our excavation drawings. However, we must not jump ahead of ourselves. Although we have permission from the owner of the site (Mike is as keen as we are to start digging), there is one major legal hurdle that needs to be negotiated.
Scheduled Monument Consent Any work that has the potential to disturb or alter a Scheduled Ancient Monument (SMC) such as Hen Caerwys legally requires Scheduled Monument Consent from Cadw. Archaeological excavation is no exception to this rule as it effectively destroys the buried deposits that it examines.
Applicants for SMC need to convince Cadw of several things. Firstly, that the excavation is adequately staffed by suitably qualified and experienced archaeologists. Secondly, that evidence can be provided of adequate resources to complete the fieldwork, finds processing, post excavation analyses and reporting to published standards and guidelines and within a reasonable timescale. Assuming that staffing, resources are adequate, the details of the project itself are crucial to reaching a decision – do the benefits in terms of information gathered outweigh the destruction incurred by excavation? Factors such as the number, size and location of trenches, sampling strategies, excavation method and day-to-day practicalities such as access and safety can all determine the success – or otherwise – of a project in achieving its objectives and can often be the subject of detailed negotiations. Indeed, if and by the time SMC is granted, many excavations are modified from the original application to a greater or lesser degree.
Normally, the task of granting or denying consent falls to the Regional Inspector of Ancient Monuments but in this case I am too closely involved in the project to make an objective decision, so the application is being considered by colleagues who cover a different region. By the time we post our next entry, our proposal will have been scrutinised and conditions set and we will tell you exactly what we will be doing this July and why.
In the meantime, don't hestitate to contact me or Bob if you have any comments or queries.
If you are interested in volunteering or just coming along for a look, the excavation will be running from the 16th to 28th of July with open days on the weekend of the 21st-22nd and Saturday 28th. Please note that in order to plan more effectively and avoid being overrun or undersubscribed we will be operating an informal booking system for volunteers this year. Please contact me for details at
Will.Davies@Wales.gsi.gov.uk or telephone 07792612246. As ever, myself and Bob are happy to answer any comments or queries on the content of the diary
With Scheduled Monument Consent granted we can tell you something about our forthcoming second season.
It is worth stressing from the outset that this excavation is better described as an evaluation. We will be excavating by hand to the topmost archaeological layer followed by small-scale sampling of specific deposits and features in order to gain structural, artefactual and hopefully dating evidence. In combination with the information retrieved from the remains of the 1960s archive, successive field surveys and our ongoing fieldwork and historical research, this approach will hopefully enable us to better understand the development of the site whilst causing minimal disturbance to the archaeological remains.
This season we will be revisiting two areas and opening a new trench in a third. You can look at the 2011 dig diary to refresh your memories of last years’ progress:
Trench A – This will entail a few days work reopening and completing the excavation across the NE angle of the large embanked enclosure. The main target is to investigate a buried soil horizon sealed under the tumbled stone of the enclosure bank / wall, which has the potential to contain dateable material.
Trench B was opened over a platform assumed to be the elusive House 2 of the 1960s excavations in order to confirm the location of a second trench noted in the 1967 interim report and alluded to in the excavators’ surviving notes. Following somewhat ambiguous results last year, when time was an issue, the trench will be reopened and expanded to examine a larger area of the structure and to. Being fully aware of the large numbers of finds unearthed by Leach and Pennant Williams on the neighbouring platform in the 1960s, excavation of the internal deposits will be limited to small samples assess the nature and preservation of any internal deposits, perhaps obtaining some dating evidence in the process.
Trench C will be located in the smaller square enclosure alongside that examined by Trench A. A long evaluation trench will cross the foundations of the three-celled internal building and a possible second structure to the north in order to ascertain their layout. Limited sampling of upper deposits will hopefully provide some dating evidence for the complex. Referring back to Bob’s post of November 21st 2011, the chronological relationship between this possible enclosed farmstead and the platform buildings to the south is a key one in our understanding of the development of the site and with broader implications for the study of medieval Welsh settlement.