Month: May 2026

The Tower House Problem

Dr Sarah Kerr, Lecturer in Archaeology at University College Cork and expert grants assessor for the Trust, takes a look at her new focus of study, Tower Houses in Ireland, and the issues involved in researching them.

Tower houses have received considerable attention in recent years, most notably Victoria L McAlister’s monograph on the subject (McAlister 2019). Her work brought to the fore the multiplicity of functions performed by tower houses particularly their crucial role in exploiting the natural environment and contributing to national and international economies. Breen and Raven focused on Gaelic-maintained areas on the long western coast of Ireland and considered late medieval tower houses as the primary expression of secular lordship (Breen et al. 2017, 151). Both important pieces of work drew upon – and contributed to – the landscape turn in castle studies by exploring the wider, often maritime, environment.

Subscribe to our quarterly newsletter

Despite these recent developments there are arguably a number of issues which inhibit our understanding of tower houses and may even restrict future avenues of research. Summarised as the tower house problem, these issues are highlighted in the hope that greater awareness may assist averting the limitations they impose. There was a vast – albeit unknown – number of tower houses in use contemporaneously in late medieval Ireland. Harold Leask, (d. 1964) the first Inspector of National Monuments in Ireland, calculated that some three thousand tower houses were documented in the Ordnance Survey maps. Therefore he provided the minimum number of how many existed prior to the mid-nineteenth century (Leask 1941; McAlister 2019, 12; Sherlock 2011, 115). Estimations have varied to over 7000 while the Archaeological Survey of Ireland list approximately 1300 known examples (Sherlock 2011, 115; Barry 1993; Barry 2008b, 129). The result is descriptions of Ireland as ‘the most castellated country in Europe’ (Barry 2008) or ‘one of the most castellated parts’ of the Irish and British Isles (Cairns 1987, 31; O’Conor 1998, 25). Certain areas represent this clearly, particularly in Munster and Leinster, with clusters of tower houses in close proximity to one another.

This vast archaeological record has posed a problem for archaeologists and much work of the 1990s and early 2000s focused on creating types and sub-types to dissect the record into manageable portions. Creation of types based on architectural similarities is useful for understanding individual examples and provincial trends but it doesn’t address the phenomenon of the tower house, the vast numbers and clustering in certain areas. As a method it is therefore limited in its usefulness. In addition, it can led to creating boxes and trying to make examples fit therein, which has obvious shortcomings and may result in overlooking the uniqueness of each example.

Subscribe to our quarterly newsletter

There is a problem more fundamental to tower house research that needs addressed: should they be considered castles? Tower houses are often referred to as castles, small castles or a type of castle and this seems rational as some examples were described as castellum, cúirt or fortalicium by their contemporaries (McAlister 2019, 15). Their architecture supports this classification, as they display features found frequently at larger castles, such as those built by the Anglo-Normans in Ireland around two centuries before tower house building commenced. These accepted castle attributes include machicolations, battlements, arrow loop windows, narrow entrances and thick walls. Other common castle features identified less commonly at tower houses include yetts (iron gate), hagioscopes (squint at the door allowing a line of sight outside), bawns (enclosure wall), adjoined towers, and some ornamentation such as hood mouldings.  

Figure 1: Leamcon (Blackcastle) Castle, copyright Dr Sarah Kerr

This is where the comparison ends, however, as the form of tower houses was distinct from Anglo-Norman castles. Tower houses were comparatively narrow towers comprising as least three storeys, such as Leamcon locally known as Blackcastle (Figure 1) in West Cork. A batter was required to aid stability (thicker walls at the base) which created slightly larger rooms at the upper levels than those on the lower levels. Windows increased in size with height, with small and sparse windows lower on the façade to create a solid and strong base. Vaults were invariably included within to further aid their structural integrity. These were often at ground-floor level but vaults on other floors were also commonplace. At the upper levels there was a wall-walk, pitched wooden roof and crenelated parapet. The latter could take the form of stepped merlons, often attributed to the Gaelic founders, or the square-crenels type, built primarily by English and Scottish settlers. In addition to the clear architectural differences between earlier stone-built castles of the Anglo-Norman period and tower house, their functions were different. Anglo-Norman castles were tools in colonialism (for example, Trim in Figure 2) and displays of the same whereas tower houses were built by Anglo-Irish and Gaelic elites, merchants and rural landowners, lay and ecclesiastical lords during a period of relative peace and economic prosperity. Some comparison between the types have focused on the towers (donjons or keeps) that were the focal point of Anglo-Norman castles however conclusions are limited due to the scarcity of work carried out on tower house settlements; that is, in their immediate surrounds.

Figure 2: Trim Castle, copyright Dr Sarah Kerr

I have recently made the argument that, combined, the above issues create the tower house problem and awareness of this is required to continue studying these buildings in an effective way. The tower house problem is typified in West Cork, a culturally distinct area west of Cork City that encompasses the south-west coastline of Ireland and comprises large peninsulas with numerous small inlets and embayments. The area is dense with extant and lost late medieval tower houses; Figure 3, is a photograph taken from Rincolisky (Whitehall) tower house with Kilcoe tower house visible across the bay. Recent fieldwork sought to examine the density of tower houses, examples of which are sometimes referred to as the ‘West Cork type’, and consider the phenomenon through the lens of spatial analysis, placename evidence and architectural analysis – results forthcoming. The fieldwork however has presented broader questions about whether we should consider tower houses castles. Either way, to do so or not should be justified perhaps with reflection upon terminological limitations. Accepting the distinctiveness of tower houses might be more productive in further understanding the reasons behind their construction and use – and crucially overcoming the tower house problem.  

Figure 3: Whitehall and Kilcoe Tower Houses, copyright Dr Sarah Kerr

Subscribe to our quarterly newsletter

Barry, T. 1993. The archaeology of the tower house in later medieval Ireland. In H. Andersson & J. Weinberg (eds) The Study of Medieval Archaeology, 211–18. Stockholm: Almqvist & Wiksell International

Barry, T. 2008a. The study of medieval Irish castle: a bibliographic survey. Proceedings of the Royal Irish Academy: Archaeology, Culture, History, Literature 108 (C): p.115-136.

Barry, T. 2008b. The study of medieval Irish castles: a bibliographic survey. Proceedings of the Royal Irish Academy. Section C: Archaeology, Celtic Studies, History, Linguistics, Literature 108C: p.115–136.

Breen, C., & Raven, J. 2017. Maritime Lordship in Late-Medieval Gaelic Ireland. Medieval Archaeology 61(1): p.149–182.

Cairns, C.T. 1987. Irish tower houses: A Co. Tipperary case study. Dublin: Trinity College Dublin.

Leask, H.G. 1941. Irish castles and castellated houses Revised edition. Dundalk: Dundalgan Press.

McAlister, V. 2019. The Irish tower house: Society, economy and environment, c. 1300-1650 1st edition. Manchester (GB): P648 – MANCHESTER UNIV PR.

O’Conor, K.D. 1998. The Archaeology of Medieval Rural Settlement in Ireland. Dublin: Royal Irish Academy.

Sherlock, R. 2011. The evolution of the Irish tower-house as a domestic space. Proceedings of the Royal Irish Academy. Section C: Archaeology, Celtic Studies, History, Linguistics, Literature 111C: p.115–140.

Cockermouth Castle Building Survey Diary – Update 2

Project Lead Tom Addyman gives an update on how the survey of Cockermouth Castle’s inner ward is progressing following the team’s second visit in April

The survey team returned to Cockermouth in mid-April to carry out the follow-on stage of the analytical survey of the inner ward.  The processed base photogrammetric data from the first survey visit was marked up with analytical data such as construction breaks and other interpretative information.  The opportunity was taken to extend survey coverage to other areas such as the mural chambers within the gatehouse and to re-survey parts of the exterior that had been better revealed following continuing trimming of obscuring ivy.

Subscribe to our quarterly newsletter

A general review was made of the complexities of the fabric, paying particular attention to structural discontinuities and changes in masonry coursing, character, stone type and alignment.  The site assessment particularly benefited from the expert insights of Dr Edward Impey and Prof Richard Oram.  Significant progress was made on understanding the evolutionary history of the inner ward buildings:

Interior of the south curtain wall showing relieving arch – possibly bridging unstable ground relating to a pre-existing defensive ditch Copyright Tom Addyman
  • Evidence supporting the idea that the gatehouse range occupied the site of a pre-existing ditch in the form of a broad structural relieving arch within the south curtain
  • Confirmation that the eastern gatehouse range incorporated the part-built remains of a predecessor of likely mid-late 14th century date that featured evidence for an intended 3-chamber rib-vaulted lower storey.  Construction resumed to a somewhat modified design within a decade or two.
Interior of the gatehouse looking east to the pend copyright Tom Addyman
  • The architectural peculiarities of the gatehouse itself were better understood, including the possibility of a draw-bridge pit within the pend, details of the pend’s octopartite vault that was found to feature decorative collars and bosses at its main junctions, and the relationship of the large vestibule chamber within to the lodging ranges to either side to which direct access was provided.
  • The narrow Bell Tower, located at the south-east angle of the inner ward, was concluded to be of early date, possibly of the early 13th century
  • The south curtain was found to incorporate many finely-worked, diagonally-tooled blocks of a predecessor building, with a number of examples displaying a narrow sunken roll of keeled profile, perhaps c.1200.  These provide the earliest date for the construction of the curtain – perhaps in the early-mid 13th century.
  • The south curtain also provided evidence for a former range of two stories with lean-to roof built against it – joist sockets and corbels, an inserted fireplace at first floor level, and indications of cross–walls.  The structure had likely been narrow and at the upper level may have provided gallery-like access between the private apartments of the hall range and the lodging block.
Interior of the lodging block to the south side of the gatehouse, looking east – showing the outline of the intended vaulted cellar. Copyright Tom Addyman

Subscribe to our quarterly newsletter

The work was coordinated with the Cockermouth Castle volunteer archive group, overseen by Alison McCann, Consultant Archivist to the Leconfield Estates.  Review of archival materials relating to the castle identified a number of significant items such as estate drawings for the various castle buildings, inventories relating to the later 17th century occupation of the site, and a number of early views including some depicting the excavation of the infilled cellarage of the inner ward in the mid-19th century.  Tom Addyman presented the ongoing work of the CST-funded survey to the archive group who were very positive in their response.

In May, the upgraded photogrammetric record is being digitised to form a line drawing set that will then be enhanced with analytical and phasing data.

Subscribe to our quarterly newsletter

You can read the first update here: Cockermouth Castle – Building Survey Diary Stage 1 – Update – Castle Studies Trust Blog

Edlingham – the 13th–14th century Northumbrian house of a well-travelled knight of the royal household

Between 1978-1982 Graham Fairclough led the excavation of Edlingham Castle, Northumberland. Rather than write up a traditional excavation report in his new book he takes a different and fascinating approach to understanding more about this castle.

Subscribe to our quarterly newsletter

Edlingham Castle, in Northumberland between Rothbury and Alnwick, comprises the remains of a medieval fortified house excavated for the predecessor of English Heritage between 1978 and 1982, and freely open to the public from the mid-1980s. The earliest known building is a hall-house (to use a slightly contested term) of c.1295-1300, probably originally inside a palisaded moated enclosure, but to this was added a chamber tower in the 1330s or ‘40s and a stone gatehouse. Its enclosing walls were rebuilt on several occasions until by the sixteenth century the place resembled a small courtyard house, which was finally abandoned by the 1650s. Edlingham thus shows in microcosm the impact of Anglo-Scottish border warfare in the fourteenth and fifteenth centuries, and of endemic lawlessness in the sixteenth, centuries. Another narrative, however, from a sociological perspective, first places the site and its owners in a national or even European context, and only in its later centuries into a more local, county-scale, world. A third Edlingham story would take most account of its architecture, notably the distinctive hall-house which before excavation was only guessed-at, and its imposing later tower that has usually been seen as part of the border ‘pele tower’ narrative – but both demand wider notice.

Air photo (with help from RAF Boulmer and one of their air-sea rescue helicopters) showing Edlingham Castle during excavation in 1979, from left, tower, hall-house, courtyard and walls of vaious dates (half-unexcavated) and gatehouse (unexcavated)

The excavations of 1978-82 focussed on the buried material remains of the site, which were described archaeologically and architecturally in summary interim articles prepared during the excavation and shortly afterwards. In my recent book, I took a different approach, and used biography – predominantly that of the castle’s first builder William de Felton, but also at some periods his peers, neighbours and family, as far as they are known – to inquire into Edlingham’s context and origins, and what might have influenced William in building the hall-house at Edlingham.

Subscribe to our quarterly newsletter

Born and probably raised in Shropshire, in some uncertain way an offshoot of the large Lestrange clan of Shropshire and Norfolk, William de Felton’s first small landholding was in Northamptonshire; he acquired land in Northumberland in the late 1270s on marriage to a Northumbrian heiress, Constance de Pontop, and he acquired Edlingham only in 1295. By then he had been a member of the royal household since c. 1278, as a personal usher (husser) in Edward I’s private chamber. He travelled the length and breadth of England and Wales with the king, spending three years in France, mainly in Gascony and Béarn. His duties included supervising the king’s personal building works, but he was also a soldier, and had fought in Wales, Flanders and Scotland. He was constable of a succession of royal castles in Wales, Scotland and England, notably, in company with the famous James of St George, Beaumaris and Linlithgow while they were under construction. Given his travels and experience, therefore, it is not surprising that whatever William was to build at Edlingham was not necessarily going to have a regional or local inspiration.

The valley of the Edlingham Burn: village, church and the castle showing its valley bottom location, copyright Graham Fairclough

The biographical approach taken in ‘A Medieval Life’ illuminates Edlingham’s origins in several ways. It is a current maxim in castle studies that castles and other elite houses should be considered in the context of their territories or landscapes, but this involves more than land and its affordances but more widely the social affinities, perceptions and abilities of their builders and inhabitants. Knowing a little of the events and travels of William’s life thus becomes significant in how we see Edlingham. My book therefore reflects on the accuracy or desirability of seeing Edlingham as a ‘Northern’ castle, coloured by a perception of Northumberland as a violent, vulnerable and distant (indeed different) border region. For William (and probably for his son and at least his first grandson), the house at Edlingham belonged to a much wider psychological and emotional landscape. Understanding the origins and early decades of the castle’s life reveals the mentality of a social and geographical network far beyond Northumberland.

Finally, let us not forget modern biography. The excavation and my early work on it occurred when I was an Inspector of Ancient Monuments, my work focussed on the material remains of the past. Since, roughly speaking, the middle 1990s, however, my work turned increasingly towards landscape ways of seeing the past. ‘A Medieval Life’, therefore, as well as being a biography, and an archaeology and history book, is also at least in intention a landscape book.

Subscribe to our quarterly newsletter

A Medieval Life: William de Felton and Edlingham Castle, 1260-1327 was published by Windgather Press (Oxbow books) in March 2025. A paper for Archaeologia Aeliana is also in preparation, and on 30 September 2026, Graham Fairclough is scheduled to give a talk about Edlingham and the book to the Society of Antiquaries of Newcastle upon Tyne.

Graham Fairclough, latterly of Newcastle University and before that Historic England and its predecessors, https://www.ncl.ac.uk/hca/people/profile/grahamfairclough.html, led five years of excavation at Edlingham Castle in Northumberland between 1978 and 1982. Early full publication of excavation results is a goal much-desired, but there is an argument for longer periods of reflection, and the recent book ‘A Medieval Life’ is a more matured discussion of the castle and the excavation framed primarily as a biography of its first builder, William de Felton.

Ten days digging at Knepp Castle

From 20 April to 29 April, a community excavation took place at Knepp Castle involving local history and archaeology groups. Richard Nevell looks back on the dig.

I woke up early on Monday 20 April to travel to Knepp Castle, navigating my way before rush hour began. I arrived on site just past 9am and suddenly it was happening. This was the first day of excavations at Knepp Castle, and with a weather forecast suggesting we would get the full use of the planned ten days of digging. More than 50 people volunteered to participate in the dig, joining from Shipley History Society, Horsham District Archaeology Group, and Worthing Archaeological Society as well as volunteers from the Knepp Estate – all under the expert supervision of Chris Butler Archaeological Services (CBAS).

Subscribe to our quarterly newsletter

The dig from afar. Photo by Richard Nevell, CC-BY-SA 4.0.

The focus of the ten-day stint was a 5m by 4m trench immediately west of the lone standing wall at the castle. There are a lot of assumptions about Knepp Castle and this was an opportunity to put them to the test, and to investigate an anomaly from the 2022 geophysical survey. The alignment was suggestive, but until we got down to the level of the anomaly what it represented was speculative.

In a broad sweep, Knepp Castle was first explicitly documented in 1210 but is likely older. It hosted several kings of England; featured in King John’s war with the barons; and was dismantled in the post-medieval period. The ruined wall stands tall in the landscape on its mound.

The first step was to remove the top soil. In an incredible feat for English weather, we not only had ten dry and (mostly) sunny days to dig but the weather in the lead-in to the dig has been mostly dry. Most of the time, rain slows things down but a sprinkling on the first day to soften the ground might have helped. That first day was the most physical, with volunteers wielding mattocks to remove the baked dry top layer of the ground. Already, artefacts started emerging – distinctly modern with some glass that might have come from beer bottles and a substantial metal piece of farming equipment which made a very effective paperweight when the weather got blustery.

Troweling began on day 2. Photo by Richard Nevell, CC-BY-SA 4.0.

As we worked our way through the subsoil, along with more animal bones and pottery we started to encounter a lot of oyster shells, mortar, and pieces of sandstone. The interpretation of these finds is pending, but oysters would have been a readily accessible food source.

Bags of oysters shells, this was’t even the whole lot. Photo by Richard Nevell, CC-BY-SA 4.0.

It is possible that the mortar and sandstone are left over from when the castle was dismantled, possibly in the 18th century. While I’m most interested in the castle’s medieval history, one of my favourite finds was a small fragment of teacup handle. That small piece of ceramic hints at people coming to the castle for a day out, and when the sun shone it was easy to see why. And though I’m no pottery expert, it was instantly recognisable.

A fragment of teacup handle. Photo by Richard Nevell, CC-BY-SA 4.0.

Over the course of the next week and a half, volunteers progressed deeper, uncovering more evidence of human activity at Knepp. The medieval pottery coming through and butchered animal bones were evidence of high-status feasting. Nails and stone may have been remnants of buildings at the castle that no longer stand.

The south side of the trench, with a stoney layer visible below the wheelbarrow. Photo by Richard Nevell, CC-BY-SA 4.0..

An area of burning mixed with slag suggests that some metal objects were created on-site. What we found was the waste from the process rather than the production site itself, but it would likely have been nearby. A volunteer asked me if this had been picked up by the geophysical survey. The magnetometry survey used in 2021 is typically good at picking up metalwork and evidence of burning, but we were digging in an area where there was too much interference from the lightning rod attached to the standing ruin to be able to pick out any detail from that survey.

A selection of animal bone, slag from metalworking, and ceramic building material found during the dig. Photo by Richard Nevell, CC-BY-SA 4.0.

As digging progressed there were regular questions about what depth we expected to find the anomaly at, and how far there was to go. When we got to the right depth we were faced with a bit of a puzzle. There was a change in soil discernable, and it looked like the anomaly may have been redeposited ‘natural’ – natural is the level at which there is no more human activity.

So what is the anomaly? The hypothesis that it might have been the wall of a building seems much less likely now. That is part of the process of putting theories to the test. At the same time, the process of excavating the castle has provided a lot of information about how it was used. Beyond the medieval period there were some prehistoric finds, pushing the history of human activity on the mound further back than anticipated. There was even a flint arrowhead, which I’ve been told is likely to date from the Early Bronze Age. The mound was a prominent feature in the landscape, and as such seems to have been a natural gathering point for people for millennia.

With the trench backfilled, the finds have been sent to the CBAS office for processing. Specialists are involved in evaluating the materials and refining preliminary interpretations. Once complete, the report will be available on the Castle Studies Trust website. In the meantime, if you’d like to relive the dig as it happened, I recorded some video updates from Knepp.

Subscribe to our quarterly newsletter