Category: Uncategorized

Picton Castle: baronial innovation in southwest Wales

In September 2023, author, Neil Ludlow, with Phil Poucher of Heneb – Dyfed Archaeology (formerly Dyfed Archaeological Trust) carried out the first modern detailed survey of Picton Castle, Pembrokeshire, funded by the Castle Studies Trust. Neil Ludlow looks at what they found in this unique and enigmatic building.

Picton Castle in Pembrokeshire has long been something of an enigma. It has a unique layout – there’s no other castle quite like it – which has been much discussed, resulting in rather more questions than answers. And it’s been continually occupied since it was built, so it’s seen a lot of alteration. While outwardly it retains much of its medieval flavour, the interiors were extensively made over during the eighteenth century so that it now presents itself first and foremost as a Georgian country seat. But beneath this veneer, much medieval work still survives – though a lot of it is tucked away behind stud-walls, in cupboards, or is otherwise obscured. Yet no structured archaeological survey of the castle had been undertaken, while the documentary record for its medieval development is more or less non-existent. We don’t even know its precise date – it’s long been attributed to Sir John Wogan, an important official in Crown service and Justiciar of Ireland 1295-1313, but being baronial work it’s unlike Crown work where accounts usually survive. 

Subscribe to our quarterly newsletter

To try and resolve some of its many mysteries, the Castle Studies Trust generously funded survey, recording and research at the castle during 2023, which was carried out by the author, Neil Ludlow, with Phil Poucher of Heneb – Dyfed Archaeology (formerly Dyfed Archaeological Trust). A full photographic record was made, along with a full 3-D survey using a Leica RTC360 laser scanner. This was not without its challenges. The castle is still occupied, as the administrative hub for the Picton Castle Trust, which means that many areas are busy, working spaces, while others are used for storage – and nearly all of it is furnished.

Longitudinal section through Picton Castle, facing north (roughly along midline) – conjectured medieval, copyright Heneb – Dyfed Archaeology
Longitudinal section through Picton Castle, facing north (roughly along midline) – as today, copyright Heneb – Dyfed Archaeology

Picton’s unique layout makes it a castle of great importance. Most castles have at least some close parallels, but Picton is effectively one of a kind. In essence, it is a towered hall-block, inviting comparisons with castles like Nunney in Somerset, and the ‘towered keeps’ of thirteenth-century Ireland, for example at Carlow and Ferns. However, close study shows that it resolves as a central first-floor hall, flanked by services and a chamber-block to form a very early example of the three-unit ‘H-plan’ house. Here, though, the end units are processed out as D-shaped towers, two on each side wall. A terminal twin-towered gatehouse lies opposite a D-shaped tower formerly lying at the western apex – seven towers in all. The hall is open to the roof; the towers have polygonal interiors (two of them disguised beneath later fittings), and contain three storeys. The ground floor is mostly rib-vaulted. The gatehouse – unusual in buildings of this kind – led onto an equally unusual ‘grand stairway’ to the hall; a second ground-floor entry probably led to an external kitchen and bakehouse. Though very forward-looking in its layout, the castle belongs stylistically to the first two decades of the fourteenth century, and analysis of the sources suggests that it was most likely built by John Wogan between around 1315 and 1320.

Picton Castle serving hatches, copyright Neil Ludlow

The castle’s spatial disposition, access and circulation are meticulously planned, while the domestic appointments show a remarkable level of sophistication for the period, including what appear to be vertical serving-hatches between the ground floor and the service rooms above. At second-floor level, the east towers and gatehouse form two integrated suites of residential apartments either side of a chapel, in a manner firmly rooted within royal planning. The opposite pair of towers, at the west end, seem to have been united internally to form a residential chamber-block, for Wogan’s officials and guests, possibly served by latrines in the former west tower; the present partition walls are later.  

Some aspects of the layout may show influence from northern Britain, or perhaps even Plantagenet Gascony. Detail shows influence from the castles of Gilbert de Clare, including the form of the spur-buttresses, the rib-vaulting and the arrow-loops. Execution of the design is however largely regional, showing great ‘plasticity’ of form and extensive squinching. There is surviving evidence for neither a defensive ditch, nor a surrounding wall until the seventeenth century, though an accompanying enclosure – containing the kitchen and other ancillary buildings – is likely from the first.

Picton Castle Great Hall area today from eastern / gatehouse end of the castle, including Georgian features

Beginning in around 1700, and spanning over 50 years, extensive works transformed the castle into an elegant country house, with magnificent and well-preserved Georgian interiors that include what seems to be only the second circular library to be built in Britain. Later campaigns included the addition of further wings and ranges. Work continued into the later twentieth century, with an extensive refurbishment in the 1960s. But the earlier work has largely survived, making the castle – along with its gardens – a popular visitor attraction.

Subscribe to our quarterly newsletter

The Lodging Ranges of Late Medieval Great Houses

Dr Sarah Kerr of University College Cork explores the importance of lodging ranges in late medieval Great Houses.

With the publication of my new monograph on late medieval lodging ranges, this post summaries what makes these buildings so vital to our understanding of the great houses in England and Wales, and calls for greater centring of function over fundamental use in buildings archaeology.

Subscribe to our quarterly newsletter

Lodging ranges are a type of collective-living building, built as part of late medieval great houses, sometimes called castles, to accommodate middle- to high-ranking members of the household. They were often long ranges occupying the side of a courtyard, divided into small yet grand rooms, these sometimes divided further into what may be an office and bedchamber.[1] Each room displays the architectural language of a comfortable living space with windows, a fireplace, and a garderobe. Each room also had an individual door that led from the courtyard or a lobby into that room and that room only, making the room a termination point (Figure 1). The individual door was almost unique in the late medieval great house, thus it sets lodging ranges apart from other types of accommodation and indeed is one of the most illuminating features in the analysis of function and use. Each room was low-occupancy, possibly single-occupancy for some periods, indicating that they were some of the most socially separated spaces in the great house with suggestions of privacy that would not become commonplace for centuries. This combination of architectural detail and separated space created some of the finest rooms within late medieval accommodation, drawing comparison with the suites of those at the head of the household.

Figure 1: Dartington Hall, Devon, Lodging Range Entrance, copyright Sarah Kerr

A dominant research focus within buildings archaeology has been, until only recently, on a building’s use which has lead to the prioritisation of practical use over more conceptual functions. Function and use are not synonymous, nor dichotomous, rather each is a coexisting layer of meaning.[2] Consider the lodging ranges’ individual doors. The door, as all doors, provided and restricted access: this was its use. The function of the door, however, was to epitomise the difference between who could enter and who could not. Its function was to communicate a sense of ownership of, or rights over, the room on the other side. Therefore there is a distinction, at least to us in the present translating the past, between the use, as a point of entrance, and the function, of communicating these meanings.

Figure 2: Dartington Hall, Devon, Lodging Range copyright Sarah Kerr

By focusing on function, the analysis reveals lodging ranges’ layers of meanings. Lodging ranges display early use of uniform elevations created through equal spacing, symmetry, and repeated features such as doors, possibly most striking at the 80m-long range at Dartington Hall, Devon (Figure 2). This strong architectural uniformity created a sense of sameness that can be extended from the fabric to the occupants, suggesting the ranges were built to construct a sense of collective identity between people drawn from different families. Within this architectural display of collective identity, however, there were expressions of individualism. Variations between rooms, both in size and provisions, created a sub-hierarchy within the lodging ranges, harking to a household in itself. This is seen clearly at Gainsborough Old Hall, Lincolnshire, which comprised three different room types concealed behind a uniform façade (Figure 3). In other examples, the differences were expressed explicitly, through larger or more intricately decorated doors and windows, ensuring the hierarchy was displayed outwardly. Centring the exploration of lodging ranges’ possible functions allows the suggestion that they were primarily built to reflect – or instil – certain identities upon the occupants. At a time of increased social mobility it could be suggested that lodging ranges were constructed in an effort to halt any aspirations of social ascension or even set in stone, quite literally, the social identities of those within the household.

Figure 3: Gainsborough Old Hall, Lincolnshire copyright Sarah Kerr

Lodging ranges draw comparison with inns, vicarial and collegiate ranges, and to a lesser extent chantry houses,[3] due to architectural similarities of individual or low-occupancy small rooms, courtyard plans, high-status provisions and uniform elevations. This allows the suggestion of a broader typology of collective-living buildings emerging in the late medieval period with collective an important distinction from communal: these were not high-density rooms comparable to dormitories.[4] While there are limitations with our over-reliance on typologies – a topic for another post – we can tentatively suggest that the collective-living building type demonstrates development from dormitories to rooms within ranges across types of auxiliary late medieval accommodation. This emergence of a collective-living building type reveals how late medieval society was responding to the shift towards specialised use of rooms and more private spaces.

Subscribe to our quarterly newsletter

The first book on late medieval lodging ranges, Late Medieval Lodging Ranges: The Architecture of Identity, Power and Space, is available to order from by Boydell and Brewer. Use code BB135 for 35% off.

Link to book: https://boydellandbrewer.com/9781783277575/late-medieval-lodging-ranges/#generate-pdf

[1] For earlier discussions on loading ranges see within the following texts: William Pantin, ‘Chantry Priests’ Houses and other Medieval Lodgings’, Medieval Archaeology, 3 (1959), 216-58 (free access); Margaret Wood, The English Medieval House (New York, 1965); Anthony Emery, Greater Medieval Houses of England and Wales 1300-1500 (3 Vols, Cambridge, 2000; 2000; 2006); Jane Grenville, Medieval Housing (London, 1997); Michael W. Thompson, ‘The Architectural Context of Gainsborough Old Hall’, in: P. Lindley (ed.), Gainsborough Old Hall (Lincoln, 1991), pp. 13-20; Christopher Woolgar, The Great Household in Late Medieval England (New Haven, 1999).

[2] Amos Rapoport, The Meaning of the Built Environment (Tucson, 1990), p. 187; Graham Fairclough, ‘Meaningful constructions – spatial and functional analysis of medieval buildings’, Antiquity, 66:251 (1992), 348-66, p. 351.

[3] Chantry houses are compared in: Margaret Wood, The English Medieval House (New York, 1965); William Pantin, ‘Chantry Priests’ Houses and other Medieval Lodgings’, Medieval Archaeology, 3 (1959), 216-58.

[4] Sarah Kerr, ‘Collective living and individual identities in late medieval England,’ Archaeological Journal, 177 (2020), 83-98.

Uncovering the Secrets of Dunoon Castle: A Collaborative Archaeological Endeavour

Harriet Tatton, Programme Coordinator, DigVentures looks at the community focused geophysical survey of Dunoon Castle in Scotland.

In summer 2023, a grant from the Castle Studies Trust facilitated an exciting collaborative venture to reveal some of the hidden mysteries of Dunoon Castle, Cowal. DigVentures, in partnership with Argyll and Bute Council, the Dunoon Area Alliance, and Castle House Museum, designed a community-orientated endeavour to learn more about the site of Dunoon Castle and raise awareness of the heritage of the Castle House Park and Gardens.

Subscribe to our quarterly newsletter

The medieval stronghold, a scheduled monument, has long been shrouded in the enigma of its past. Thanks to the support of the Castle Studies Trust, a comprehensive geophysical survey and community activity program were initiated, providing a fascinating glimpse into the castle’s rich history.

Community participants getting to grips with geophysics in front of the Castle mound © DigVentures

Professional archaeologists, building historians and amateur enthusiasts from the local community came together to conduct a non-invasive survey that aimed to unearth the secrets of Dunoon Castle and its surrounding parkland. The results not only contribute to the ongoing heritage management of the site but also enhance the narrative that can be shared by local heritage organisations such as the Castle House Museum. In total, just under 100 people were involved in the project, which included training workshops, a public talk and tours, and activities for children and young families.

Looking for overgrown walling on the top of the mound © DigVentures

Dunoon Castle has stood witness to centuries of history. Documentary evidence traces its roots back to the early 13th century AD, and today, it stands as a modified natural mound with a curtain wall, hinting at its medieval past. There is debate over the castle’s original purpose, with suggestions ranging from a motte-and-bailey castle to a ‘castle of enclosure.’ Recent interpretations lean toward the latter, aligning with a group of castles built during the 12th and 13th centuries by successive Scottish kings.

Volunteers exploring the Castle mound © DigVentures

The castle’s complex history is evident in distinct phases. In the medieval and post-medieval era, Dunoon Castle played a pivotal role in Scottish history. From its Lamont origins, through conflicts and regencies, a visit from Mary Queen of Scots, to its eventual abandonment around AD1650, the castle witnessed tumultuous times. The 19th century marked a transformative phase as Dunoon Castle became a focal point for James Ewing’s architectural ambitions. His castellated mansion, set against landscaped gardens, turned Dunoon into a sought-after holiday destination.

As the 20th century unfolded, Dunoon Castle assumed new roles during both World Wars, serving as a strategic defence point with searchlights and coastal batteries. The aftermath of war saw the castle’s hill repurposed for military activities during the Cold War, reflecting the evolving strategic importance of the site.

Geophysicist Kimberley Teale with the project’s team of community participants © DigVentures

During the weekend of activities, DigVentures’ small team of archaeologists and local people aimed to make sense of this historical jigsaw puzzle. The magnetometer survey, conducted in areas north and south of the tennis courts and south of the castle museum, uncovered intriguing details. Possible garden wall foundations around the museum, features related to landscaping of the castle gardens, and remnants of former garden plots and a fire station emerged from the data, making sense of the castle’s historical evolution.

Simultaneously, the earth resistance survey corroborated these findings across the same areas, providing a comprehensive dataset for analysis. The revelation of garden features, potential gateposts, and additional elements related to the castle’s history further enriched the understanding of Dunoon Castle’s past.

The team review the results at the end of a day of geophysical survey © DigVentures

Over the weekend, Dr. Louise Turner spearheaded a rapid historic buildings assessment of the castle remains, and although not an exhaustive survey, the assessment hinted at the Victorian origins of much of the visible walling within the castle ruins. Surprisingly, limited evidence was found for the earlier phases of the castle and raises intriguing questions about its earlier configurations; documentary evidence suggests that a castle has been present at the site since at least the 13th century, but very little of this phase appeared to be visible.

Subscribe to our quarterly newsletter

Complementing these efforts, historic building and aerial 3D photogrammetric drone surveys were conducted. These surveys not only highlighted obscured walls and earthworks but also uncovered hidden features concealed by vegetation overgrowth. A resulting topographic survey, compared against historical maps and surveys, revealed that a substantial portion of the castle walls, previously recorded, is now obscured or even considered submerged beneath the current ground level.

The findings underscore the importance of ongoing conservation efforts, as vegetation threatens to permeate the fabric of the surviving castle walls, potentially causing damage, in particular to the earlier, less visible phases of the castle’s structure. This collaborative project aligns with best practices outlined by Historic England, the Chartered Institute for Archaeologists, and the Europae Archaeologiae Consilium, ensuring the responsible and insightful exploration of Dunoon Castle’s captivating history.

As the dust settles on this archaeological venture, Dunoon Castle stands as a testament to the power of collaboration, community engagement, and cutting-edge survey techniques in unravelling the mysteries of the town’s history. DigVentures hopes to do more work at Dunoon Castle in the near future.

Subscribe to our quarterly newsletter

To look at the geophysical survey report you can go here: Geophysical surveys at Dunoon Castle | Castle Studies Trust

What use a gallery?

Dr Katherine Weikert Senior Lecturer in Early Medieval European History at the University of Winchester takes a look at galleries in Anglo-Norman keeps.

At many Anglo-Norman tower-keeps, there is a significant part of the castle which remains generally under-discussed: the gallery. Often circling above a main room or hall space at least one storey above the floor level, a gallery is normally interpreted to the public as a space for musicians, or a passageway, if noticed at all.

Subscribe to our quarterly newsletter

Part of the neglect of the gallery often has much to do with the survival of the building remains. There is considerably reduced size and physicality to these galleries compared to great rooms such as halls and chambers. One of the best-preserved galleries is in the White Tower, London, which has galleries above both the main rooms and the connected St John’s Chapel. Dover Castle (Kent), too, demonstrates this (though you must use your imagination to remove the later brick barrel-vaulted ceiling to envision it!) But many others such as at Rochester remain in a state that make it more difficult to understand. This means that the keep galleries are often neglected not only in scholarship, but in public interpretation.

Rochester Keep Interior copyright Katherine Weikert

But understanding these galleries provides new insight to the performance of prestige at royal courts and high-status halls in the Anglo-Norman period. What comes to light is the importance of seeing, and being seen. Although in most current states of preservation this is difficult to perceive, between keeps which retain their galleries in safe conditions, and studying the galleries from a point of view of pathways and viewsheds, their meanings and use become more transparent.

Bedroom in Dover Castle Keep copyright Katherine Weikert

In castles such as the White Tower, Dover, and Rochester (Kent), the galleries providing viewsheds into the grand rooms below were a part of routes which indicate that the galleries were important parts of the castle’s ‘choreographed’ space. These passageways to and from these galleries were on paths that included ‘prestige’ places such as the hall and chamber. At Dover, the views from the second-floor galleries actually overlooked both the hall and chamber – leading to even more insight to research which indicates that a chamber was not so much ‘private’ as ‘more select.’ At Rochester, the gallery on the third level equally overlooks the main rooms on either side of the spine wall on the second floor. These galleries also provide a route between the two rooms, an alternative to the two doors in the spine wall between them: a more circuitous route no doubt, but one that provided different opportunities to see the whole rooms below, and be seen above them.

Hedingham Castle Gallery Arches, copyright Katherine Weikert

More rarely, castles retain enough fabric to actively see these viewsheds. Hedingham Castle (Essex), where some of our donors enjoyed a special event in 2021, is in superb and even liveable condition. Here, it is possible to see, experience and understand these paths and views from the gallery. At Hedingham, the gallery overlooks both sides of the grand first-floor room which is divided – but not separated – by an impressive two-storey arch. Views from the gallery overlook both sides of the hall below. Although the gallery arches are undecorated – the ground-floor arches below them have chevron patterns – the size of the gallery arches directly echoes the size of the ones below them. These are wide, open arches from which a person could be seen, while they are simultaneously seeing the action in the hall.

Hedingham Castle, view of the great room from gallery, copyright Katherine Weikert

In all of these castles, people in those great rooms below the galleries had also to consider the impact of the visual message that they were sending to those above. This could include regular matters of state or court; ceremonies or feasting at important events; crown-wearing at notable events; the reception of other high-ranking aristocrats and foreign emissaries, and more beyond this. At places such as St John’s Chapel in the White Tower, the liturgical message also needed to be seen and received. Why make such a display at times and places such as these if it couldn’t be seen? Galleries provided an audience space for those watching the enacted scenes below. In some circumstances, depending on how the hall was set for a number of guests, it may have been possible to better see the action from the gallery than from the floor!

These gallery viewing points were also important places to be seen, not just to see. At Dover, the king and queen could sit on their dais and very likely see those in the galleries in front and above them better than they might see who was at the back of the single-levelled floor of the hall itself. Remaining stone decoration of castle gallery arches further helps to understand the visual impact of being seen in these places. For example, the arches in both the great hall and the gallery at Hedingham echo each other with scalloped capital. At Dover, the gallery behind the high end of the hall is large enough for an entire visual tableau to be created, framed and presented to those in the hall below, an opportunity to view and perhaps control the vision of who was being seen in the gallery behind the seated king and queen.

Hedingham Castle, south entrance to the gallery, copyright Katherine Weikert

Further surviving stonework can lend even more to the interpretations of galleries as prestige spaces. At Hedingham, the south entrance to the gallery level has a particularly fancy doorframe with double-shafted columns with a spiral pattern and beading carving. This gallery was no service level or low-status route, but one that announced its importance in stone. As in the interpretations at Conisbrough Castle (South Yorkshire) and its fancy doors leading to high-status spaces, we should envision galleries such as these as controlled spaces, possibly with doormen allowing or denying entry into them, with access routes including a series of checkpoints in order to reach them.

What use a castle gallery then? More attention needs to be paid to these parts of castles. Although no doubt a fluid space that changed use as needed – as were most castle rooms – the galleries of Anglo-Norman keeps need to be realised as places for high-status members of the court and the castle community not only to see proceedings below, but be seen. The need to be noticed in attendance to a lord cannot be underestimated in the middle ages, where status was important, and malleable. There is more to a castle gallery than meets the eye…and the visual element of them is a key part to understanding them.

Subscribe to our quarterly newsletter

Further reading:

Robert Liddiard, Castles in Context (Macclesfield: Windgather Press, 2005).

Katherine Weikert, ‘Creating a Choreographed Space: Anglo-Norman Keeps in the Twelfth Century,’ in Buildings in Society: International Studies in the Historic Era, edited by Liz Thomas and Jill Campbell (Oxford: Archaeopress, 2018), 127-40.

Berkeley Castle Donjon and Moat

Berkeley Castle Project Excavation Director Dr Stuart Prior takes a look at one of the many interesting discoveries made during the dig which is part of a new book looking at 15 years of excavation.

Between 2005 and 2019 the Berkeley Castle Project (BCP), conducted by University of Bristol, carried out excavations and survey work at Berkeley Castle, which have led to the publication of a new book. Excavations in 2015, of Trench 19, were able to gain insight into the early origins of the castle and the donjon that was constructed when the castle was built in stone by Robert FitzHarding in 1153–1154.

Subscribe to our quarterly newsletter

It was originally believed that the first stone castle erected at Berkeley comprised a circular shell keep, but the BCP was able to shed new light on this aspect of the site’s past and its architectural evolution. In a Castle Studies Group Bulletin (CSG Bulletin 18, 2014), Neil Guy suggested that the castle may have had a square or rectangular donjon or keep that may have been modified as the basis for the Thorpe Tower by Thomas [III] Berkeley (1292–1361). Trench 19 was designed to look for evidence of the north-west corner and west wall of this postulated donjon. The argument here was that Thorpe Tower was not wholly created ‘as new’ in the 14th century but was instead a part-relic structure arising from a 1340s re-modelling of the 12th century castle. Namely, two corners and one side of a square donjon which abutted the north side of the ‘motte’, and for which the shell-keep encasing the motte was an inner (and elevated or upper) bailey.

Figure 1: Plan view of Trench 19 showing heavily robbed-out building foundations copyright Berkeley Castle Project

The archaeological remains observed in Trench 19 (Fig. 1) appear to demonstrate the presence of a heavily robbed-out building with structures of two later phases overlying it (Fig. 2). The orientation of the first structural phase (contexts 1912 and 1916) and the robber trench (context 1908) associated with it is in alignment with the south-facing elevation of Thorpe Tower. This orientation suggests that this first phase was associated with, and presumably connected to, Thorpe Tower. It is probable, therefore, that context 1912 represents a heavily robbed wall which is comparable, and most likely contemporary with, wall J3, identified by the 8th Earl, who was an amateur archaeologist, which extended from the northern elevation of Thorpe Tower (TBGAS, 1927, vol.49, 183-93 & 1938, vol.60, 308-39).

Figure 2 – Location of proposed donjon overlying plan of 8th Earl’s excavations. Copyright Berkeley Castle Project

It appears then that the shell-keep and Thorpe Tower are of a single phase, most likely dating to the mid-12th century. While there is no evidence currently that contexts 1912, 1916 and wall J3 are contemporary with this primary construction phase, it must be noted that the wall (1911) overlaid context 1912 and re-used some of its stone. Further to the evidence from Trench 19, the rear wall of this fortification can still be seen, incorporated into the castle’s later form (Figs. 3 & 4).

Figure 3 – 17th century painting by Dankerts showing original height of donjon along with remnant of projecting wall (heading north towards church).
Figure 4 – Aerial view of Berkeley showing reduced height of donjon; and with addition of 18th century laundry attached to north. Copyright Berkeley Castle Project

Accompanying the donjon, there are several medieval documents that record the cutting of moats around Berkeley Castle. In The Cartulary of St Augustine’s Abbey, Bristol, an entry made between 1171 and 1190 records a grant made by Maurice de Berkeley [I] to St Augustine’s of a rent of 5s from his mill below the castle, some tithes of pannage, and common pasture for a plough team ‘pro emendatione culpe mee de fossato quod feci de cimiterio de Berchel circa castellum meum’ (charter no. 78; Walker, 1998, 46–7), which roughly translated means ‘in recompense for my offence committed upon the cemetery of Berkeley in cutting a ditch around my castle’. This suggests that Maurice cut a moat around his castle, which encroached upon part of the cemetery, and he was subsequently fined for his actions. The grant is again confirmed sometime between 1190 and 1220 by Maurice’s son, Robert [II] (charter no. 119; ibid., 69–70).

During this period then, the castle comprised an ovoid shell-keep with adjacent forebuilding, the curtain wall of the inner ward and the Norman Great Hall, all wrapped around the skeleton of the earlier motte and bailey. Excavations carried out by the 8th Earl between 1917 and 1937 (TBGAS 1938, 321) demonstrated that the shell-keep was already adequately defended by a moat that ran around its base on the southwest, north-west and north-east sides – which may have encircled the earlier motte and bailey – and records show that Maurice [I] dug a deep moat around the south-east side of the castle, presumably to complete the defensive circuit, and diverted the Newport brook and others towards the castle to fill it.

More information on the Berkeley Castle Project (BCP), on the castle itself, and on the excavations and survey work conducted by University of Bristol can be found here: https://www.archaeopress.com/Archaeopress/Products/9781803275680

Subscribe to our quarterly newsletter

Bibliography

Earl of Berkeley, 1927. Berkeley Castle. Transactions of the Bristol and Gloucestershire Archaeological Society

49, 183-193.

Earl of Berkeley, 1938. Excavations At Berkeley Castle. Transactions of the Bristol and Gloucestershire

Archaeological Society 60, 308-339.

Walker, D. 1998. The Cartulary of St Augustine Abbey, Bristol. Gloucester: Bristol and Gloucestershire Archaeological Society.

Coldridge Castle: A Wood-Smoked and Charred Landscape

Alison Norton, PhD student at Christ Church Canterbury, who is studying medieval castles and landscapes in South West England looks at one of the castles she is studying as part of her research, Coldridge in Devon.

Coldridge Castle is a motte and bailey located in Mid-Devon, roughly 16km northeast of Okehampton (Figure 1). The castle sits within dense woodland, known locally as Castle Wood, along the eastern banks of the River Taw. Past investigations of Coldridge and its community are limited to gazetteer entries due to a scarcity of text-based and archaeological source material. This evidence gap has resulted in a lack of historical and archaeological context of Coldridge as an individual site within its surrounding landscape. To provide context and understand why castle builders chose this particular site, my research first utilised place name and landscape evidence. This source material provided a framework for me to computationally generate theoretical landscape models and test siting theories using GIS and LiDAR.

Subscribe to our quarterly newsletter

Figure 1: LiDAR imagery of Coldridge Castle. Image: Alison Norton; LiDAR Data: DEFRA

Coldridge (Old English for “charcoal ridge”) indicates a pre-Conquest landscape of the production and trade of charcoal. This resource was a valuable fuel source and was likely supplied to surrounding estates, such as Crediton and Winkleigh. As a charcoal-burning site, the community would have been integral to the local economy, requiring easy access to various trade networks. In addition, given charcoal-burners heavily relied on a continuous supply of coppice, Coldridge would need to be positioned in a well-wooded landscape. When expanding my landscape analysis to include surrounding estates, place-name, and supporting Exon Domesday, evidence highlights two manors, Leigh and Loosebeare, that indicate the local area was predominantly woodland (Figure 2).

Figure 2: Theoretical boundaries of local woodland based on place name evidence. Copyright: Alison Norton

It is very likely, based on this source material, that woodland associated with Leigh (Old English for “wood”) and Loosebeare (Old English for “place that pastures pigs”) was contiguous with that of Coldridge. This wooded characteristic of the castle’s surrounding landscape incorporates ideas of seasonality regarding visibility from and of the site. My initial thoughts about this seasonal element were as follows: I thought travellers moving along local trade routes would have limited visibility of the castle. This observation could also apply to the castle having limited visibility of movement within its surrounding landscape. Second, it felt possible castle builders were primarily influenced by the need to have targeted visibility over local production.

To test these theories, I applied least-cost path and viewshed analyses. The former is a tool that predicts movement between various places of interest. It takes topographic and elevation data and generates predicted routes that exert the least amount of energy or time to travel to and from said points. The latter is a tool that analyses visibility from a specific point within the landscape. Viewsheds show what is visible and not visible to an observer and can be adjusted to include various heights and distances. For example, a viewshed can determine what is visible within a 2km radius to an observer standing atop a 9m tower. Results from these analyses revealed the castle had concentrated views over its immediate surroundings, particularly the bridge in the neighbouring manor, Brushford. In contrast, the castle’s viewshed showed sporadic visibility over predicted routeways that could indicate monitoring movement to and from Coldridge was not a primary factor in castle siting decisions. When I applied viewshed analyses from routeways and surrounding estates, results showed travellers moving east towards the castle from Brushford and Winkleigh held more targeted views of Coldridge. Further research is in progress, though it is probable visibility over the production and trade of charcoal were the primary influencers for castle builders as opposed to sweeping visibility over the landscape (Figure 3).

Figure 3: Cumulative viewshed results of the castle and its community. Copyright: Alison Norton

If any other student would like to promote their work with a short blog post on our site, the Trust would be interested in hearing from you. You can contact us at admin at castlestudiestrust.org .

Subscribe to our quarterly newsletter

Transforming our understanding of Shrewsbury Castle

With the excavation report on the third and final season of excavation which the CST has funded now published on our website, project lead Dr Nigel Baker looks at what has been achieved since the first work in 2019 to now.

Just over a century ago Shrewsbury Castle began a new phase in its long life. In 1925 its principal surviving building, having been in use as a private dwelling since the castle was finally de-munitioned in 1686, became the meeting hall of Shrewsbury Borough Council, set in extensive landscaped gardens covering the remains of the motte and inner bailey, the outer bailey having (mostly) disappeared beneath the growing town by c.1300. Shrewsbury Castle remained more or less untouched by archaeology for the remainder of the 20th century. This changed in 2019 with the award by the Castle Studies Trust of a grant for a season of geophysical survey and excavation in the inner bailey. Following permission from Shropshire Council, the site owners, and Historic England, its legal guardians, the work took place in May and July 2019, the geophysics by contractors Tiger Geo and the excavation team made up of experienced local volunteers and staff and students of University Centre Shrewsbury. The results were unexpected.

Subscribe to our quarterly newsletter

Shrewsbury Castle Excavation 2019 showing the width of the ditch around the motte using deckchairs (copyright Dr Nigel Baker)
Arrow heads found in Shrewsbury Castle Motte Ditch (copyright Dr Nigel Baker)

Immediately under the turf was natural glacial gravel: the top of the hill on which the castle had been built; the ground surface had been lowered sometime in the past, removing nearly all archaeological remains. This was almost certainly the work of the young Thomas Telford who, from 1786 to 1790, lived in and ‘restored’ the castle for its owner, Sir William Pulteney, M.P. for Shrewsbury. However, archaeological strata were found to have survived within cuts into the natural gravel, and two of these were of major significance. The first was the edge of a previously-unknown ditch around the base of the motte. Medieval cooking-pot sherds of late 11th-13th-century date were found in its lowest excavated layers, along with two armour-piercing crossbow quarrel heads. The second significant find was of a pit containing in its fill a piece of decorated bone and two types of pre-Conquest (Saxon) pottery: Stafford-type ware, distributed widely across the emerging towns of the region and already well represented in Shrewsbury; and a limestone-tempered fabric, TF41a, never before seen in Shrewsbury, which had been made in the Gloucester area and probably imported up the Severn. This confirms that there was pre-Conquest activity on the site of the castle, and, along with the Domesday evidence that there was a church of St Michael there by 1086, may point in the direction of a high-status pre-Norman presence on this tactically-significant site controlling access to the ancient borough.

Shrewsbury Castle Excavation 2020 (copyright Dr Nigel Baker)

Excavation resumed in the autumn of 2020 with a trench seeking a sample profile through the west rampart of the inner bailey. This turned out not to be medieval in date. Both the west and the north rampart were probably created as part of Thomas Telford’s landscaping work in 1786-90. But, intriguingly, below the west rampart there was no sign within the trench of the natural hilltop gravel found close by in 2019 at a depth of just a few centimetres. The explanation may be that the bailey was enlarged westwards between the Norman period and the later medieval period, by dumping soil and levelling-up behind a new curtain wall.

Shrewsbury Castle Excavation 2022 on the motte top (copyright Dr Nigel Baker

The final season of excavations took place in 2022 on the top of the motte, and outside the north curtain wall. Telford is known to have demolished ruined medieval buildings on the top of the motte and replaced them with the surviving two-storey Gothic summerhouse there. Excavation showed that Telford’s activities had, again, removed most of the archaeology but that the foundations of early medieval timber buildings (beam slots, a post pad, post holes) survived where they had been cut into the motte material. No definite trace was seen of the ‘great wooden tower’ which is documented on the motte top until its collapse in 1269-71.

New light was also shed on the motte by vegetation clearance on its south side, revealing for the first time remains of buildings incorporated in the masonry of the retaining walls. This work was undertaken on behalf of Shropshire Council for a new conservation-management plan, currently at consultation stage, which includes photogrammetric surveying of all the castle structures. This permanent stone-by-stone record not only forms the basis for the next vital stage of work – identifying and specifying long-needed repairs – it also offers new archaeological insights, including the identification of the probable primary sandstone rubble fabric of the curtain walls. This was in turn followed by some research carried out by Jason Hurst on Civil War musketry damage in 2023 (Potential shot damage at Shrewsbury Castle – Castle Studies Trust Blog) . And now, the process of publishing this body of new archaeological, architectural and historical information is just beginning…

Subscribe to our quarterly newsletter

Picton Castle – form, function and affinities – new work at an enigmatic Pembrokeshire castle

Starting on September 11 2023, Dyfed Archaeological Trust, with Neil Ludlow, will undertake survey and recording at Picton Castle, near Haverfordwest in Pembrokeshire. This work, wholly funded by the Castle Studies Trust, will be fully analytical: it will provide a comprehensive record, underpinned by new research, in an attempt to unravel some of the mysteries of one of Wales’s – and indeed Britain’s – most enigmatic castles. Here Neil Ludlow and Phil Poucher from DAT explain more about the project.

Subscribe to our quarterly newsletter

Picton Castle has been continuously occupied since it was built, and was substantially modified over the centuries, now wearing the veneer of an eighteenth-/nineteenth-century country house. Underneath, however, is a baronial castle from around 1300-20, of a highly unusual plan without close parallels in Britain and Ireland. A small, compact building, it comprised a hall-and-chamber block flanked for D-shaped towers, one near each corner, with a twin-towered gatehouse at one end – possibly unique – and a D-shaped tower, now lost, at the other. Variations on this basic ‘towered hall-block’ layout are seen in castles of similar date – and somewhat earlier – in the West Midlands, Scotland and the border, Ireland and even southwest France. But none fully mirror Picton’s plan-form – which might conceivably have been at least partly modelled on the large ‘keep-gatehouses’ of the late thirteenth century.

In addition, Picton shows strongly regional attributes including a first-floor hall, corbelled parapets and an abundance of squinched features in the external angles. The towers rise from pyramidal spur-buttresses with an unusual octagonal footprint, otherwise seen only in the gatehouses at St Briavels Castle (Gloucs.) and Tonbridge Castle (Kent). So a variety of influences – regional, national, international and purely personal – may lie behind design at Picton.

Picton Castle Ground Floor (incl ribbed vaulting)

No structured survey and analysis of the castle has yet been undertaken, and it is fundamental questions like this that the present study will address. In addition, little is known of how the castle actually functioned. The service end of the hall is currently assumed to have lain towards the gatehouse, where blocked doorways possibly led to a buttery and pantry; at the opposite end, it’s possible that the two western towers were united to form a storeyed chamber-block, to which the lost D-shaped tower was a bedchamber. But medieval access arrangements are still a mystery, as are the use and relative status of many internal spaces. For instance, it’s speculated that a broad flight of steps might have led from the gatehouse, through a ‘processional’ archway, to the hall. But the ground floor is, unusually for the region, rib-vaulted throughout – can it really have been ‘cellarage’, or did it provide access (and an anteroom) to the hall and chamber? Two spiral stairs connect ground- and first-floor level, one of them with very broad treads – were the distinguished by the status of their users? Or did the vaulted ground-floor corridor lead to a stair accessing the high end of the hall?

Not all internal walls have been dated – some, at least, may be post-medieval. Similarly, it is clear that not all medieval features such as openings, entries, stairs, latrines, fireplaces, ovens and hearths have yet been identified. Their correct identification and dating will tell us a lot more about status and usage of interior spaces, and about circulation between them. The location of the kitchen and bakehouse – and method of water-supply – are also still unknown: were they separate from the main building? A chapel was in existence by the seventeenth century, and is assumed to have a medieval predecessor, but it is yet to be shown whether it lay over the gate-passage like its successor.

The 2023 work aims to resolve such questions, and to achieve a full understanding of the form, functions and affinities of the medieval building. A combination of total-station theodolite survey, drone photography, drawn elevations, a high-resolution photographic record and, where possible, 3D modelling will be used to obtain a comprehensive record, supported by new research. All evidence for medieval features, openings and architectural detail will be recorded, along with former levels and access between them, and any indication of different building campaigns. The report, with all survey drawings and photos, will be posted on the Castle Studies Trust website. 3D models will be accessible via Sketchfab.

Subscribe to our quarterly newsletter

Contextualising Bamburgh Castle: wells, towers, mounds and more!

Dr Jo Kirton co-director of the Bamburgh Research Project takes a look at the work they are doing at Bamburgh Castle, funded by the Trust.

Throughout 2023 and early 2024 the Bamburgh Research Project will be utilising funding from the Castle Studies Trust to further explore Bamburgh Castle’s medieval outworks, particularly the area outside St Oswald’s Gate where our current excavation is underway as part of our annual field school. Our project is titled ‘Contextualising Bamburgh Castle: wells, towers, mounds and more!’

Subscribe to our quarterly newsletter

crForeground shows the area currently under excavation by the BRP with St Oswald’s Gate visible at the top of the steps and West Ward of Bamburgh Castle present in the background.eated by dji camera

Bamburgh Castle

Bamburgh Castle rises from the North Sea coast of north Northumberland. It lies on an undulating, 3.2-hectare, outcrop of dolerite bedrock that stands up to 30m above the surrounding countryside. Bede describes it as a palace site of the Anglo-Saxon kings of Northumbria from the later 6th century and it remained a royal castle until its semi-abandonment after a great siege in 1464. Rebuilt by the Lord Crewe Trust in the 18th century and again by the 1st Lord Armstrong at the end of the 19th, the fortress has long held a special place in the history and culture of the region.

The majority of the archaeological work at Bamburgh has concentrated in the low-lying West Ward at the north of the castle. A complex deeply stratified, finds-rich, site has been revealed archaeologically. By contrast, the Inner Ward of the castle, at the very top of the hill, remains a built-up area. Modest scale excavation of the accessible areas has provided an important balance, giving insight into the heart of the citadel to contrast with the occupation and industrial activities we see in the West Ward.

Aerial shot of St Oswald’s Gate with steps leading down to current excavation area.

St Oswald’s Gate

St Oswald’s Gate and the outworks beyond lie in the area of the original entrance to the castle. It is very likely that the siege castle (named Mal Voisin in the Anglo-Saxon Chronicle) was built close to this gate in AD 1095. When the main access was re-sited, the entrance here remained as an important postern, perhaps serving a small adjacent harbour immediately to the north of the site. This area now forms the BRP’s main investigative focus. The outworks consist of strong walls enclosing a trapezoidal area with the Tower of Elmund’s Well, with a more recent wall and postern to the west.

Map depicting current area of excavation (Base map Crown Copyright/database right 2022. An Ordnance Survey/Edina supplied service).

The outworks at St Oswald’s Gate are a rare case at Bamburgh as they have not been subject to extensive rebuilding in the post medieval period. Other than the reconstruction of the tower as a cottage, the outworks represent an astonishing window into mostly unaltered medieval fabric still standing at Bamburgh.

Aerial shot of Elmund’s Tower and suspected medieval well location.

Work to Date

Recent investigation by the BRP has revealed that a substantial structure still survives below ground. This is in the form of an L-shaped corridor and steps down into the room that is thought to be the tower basement that contained the well. The presence of two splayed narrow windows appears to further indicate that this is part of the medieval Elmund’s Tower. Our primary aim this year is to continue to reveal the full extent of the tower and identify any remains of the well depicted on the 19th century survey. 

Steps down into the room that is thought to be the tower basement that contained the well.
Arch into the well room of the tower

What will the Castle Studies Trust (CST) Funding be used for?

There are two primary aims for the CST funding:

  • The first is to contextualise our recent excavations at Elmund’s Tower through geophysical survey (GPR and Magnetometry) and to undertake a masonry survey of the castle’s associated extant outworks. This will include using photogrammetry to create a 3D model of the standing outworks and internal structures of Elmund’s Tower. The survey will be undertaken in conjunction with a metric survey of the structures outlines. This work is currently underway and we look forward to sharing the results with the CST. You can follow our progress on our blog: https://wordpress.com/post/bamburghresearchproject.wordpress.com
  • The second focus for the funding will be on disseminating our discoveries to the wider public. We will install signage for visitors, who cannot currently visit this area of the castle with a QR code for the 3D model, granting online access to Elmund’s Tower and the wider outworks over the winter period. This information will also be replicated and enhanced with the creation of a new webpage on the Bamburgh Castle website. Alongside these permanent additions we will continue to share our work through our blog and social media.

Subscribe to our quarterly newsletter

Dig Diary Six: Lowther Castle and Village Project Highlights

In her final blog piece (for now) Project Lead, Sophie Ambler, looks back at the three excavation at Lowther

The on-site investigation of the medieval castle and village at Lowther (Cumbria) has now drawn to a close. Over the past month, a team from Allen Archaeology, UCLan, and Lancaster University has been exploring the site through geophysical surveying, excavation and archival research. We now have a geophysical survey of the village to analyse alongside LiDAR and the original earthworks survey. Over the coming months, our small finds will be analysed, together with soil samples, in the hopes that they yield dating evidence, and a report will be prepared drawing together the results from our trenches.

Subscribe to our quarterly newsletter

Meanwhile, we have the chance to reflect on what’s been a thrilling month of investigation.

Uncovering the construction of the ringwork castle and assessing its situation has helped us to get a sense of the site’s place in the broader landscape. Since the area in which the castle stands is now wooded, this takes some imagination.

1. Lowther medieval castle OS map with label

The ringwork castle is sited on the edge of Lowther’s western escarpment, which runs down to the River Lowther. The ringwork’s positioning is clearer in OS maps (image 1), where its proximity to the river and the steepness of the escarpment are evident. Today, one can get some sense of the impact of this positioning by heading eight hundred metres or so south to take the view over the escarpment from the Jubilee Summer House, in the grounds of the nineteenth-century castle (image 2, and view the panorama on Google Maps.), although the escarpment is far less steep here. Originally, the castle would have commanded wide-ranging views to the west, across the river to Askham Fell, and would have been a highly visible – perhaps dominant – landmark for miles around.

View of western escarpment down to River Lowther from south of 19C castle

The scale and the construction of the ringwork castle has also become clearer. Again, given the tree coverage and overgrowth, it’s long been hard to perceive the earthwork’s size and form. It’s been hard also to capture the earthwork in photographs – but throughout the project Lowther’s resident photographer, Tony Rumsey, has been busy. His drone footage gives a much clearer sense of the site (image 3). To the top of the picture, the western escarpment drops steeply down from the earthwork to the river. In the foreground, Trench Two cuts into the earthwork’s northern bank.

Drone Trench 2 bank of ringwork castle in woods_Tony Rumsey

The image also shows how the floor level of the earthwork’s interior is significantly higher than the exterior ground level. As Trench Two revealed, the ringwork was constructed as a large, roughly square mound with layers of earth and stone, with its banks built up further to gird the mound. Meanwhile, the interior was topped with a metalled surface. We expect that the banks would have been surmounted by a simple fence or palisade (Trench Two did not reveal any postholes to indicate this palisade, but this is not surprising given that the top of the bank has almost certainly been lost to slippage).

Drone Trench 4 metalled surface entrance to ringwork castle Tony Rumsey

The metalled surface covering the interior of the ringwork is also clear in Trench Four (image 4). This trench takes in the approximate area of the ringwork castle’s entranceway, which cuts through the eastern bank. The entranceway may have included a wooden gateway, although we haven’t found firm evidence of one in Trench Four, and perhaps would need to open a larger area to be sure. Trench Three picked up a trackway (image 5), noted in the earthworks and geophysical surveys, which linked village to castle and brought visitors to the entranceway.

5. Drone Trench 3 trackway into castle_Tony Rumsey_

One of the highlights throughout the project has been welcoming visitors to the site. The project’s archaeology students from UCLan have been giving tours to those who’ve ventured down to the site during the course of the dig, keen to know more about what we’ve been uncovering. On Saturday 15 July, we were delighted to welcome representatives of the Castle Studies Trust and share with them our ongoing work (image 6), as well as members of several regional history and archaeology societies. We also had a visit from Professor Alice Roberts and the team from BBC2’s Digging for Britain (image 7), who plan to feature the project in their next series.

6. Visit of CST
7.Digging for Britain with project team

We still have further to go in analysing our findings and expanding our investigation of Lowther’s medieval castle and village, but are very pleased with how the on-site phase of our project has gone – not only in exploring the remains of an important medieval castle site, but also in training a new generation in castle archaeology, and encouraging public appreciation of castle studies. The project team is extremely grateful to the Castle Studies Trust for funding the project and for its support throughout, and would also like to thank the Lowther Castle and Gardens team for their help and hospitality over the past month. I’d also like to say a huge thank you to the indomitable Jim Morris, who has led the UCLan archaeology contingent, the hardworking and dedicated cohort of UCLan archaeology students, and the excellent Allen Archaeology team (Jonny Milton, Rob Evershed and Tobin Rayner).

Subscribe to our quarterly newsletter