Fig. 1. Dundrum Castle from the south. From Francis Grose ... · 38 THE CASTLE STUDIES GROUP JOURNAL NO 28: 2014-15 CSG Annual Conference - Belfast - April 2014 - Dundrum Castle Fig

THE CASTLE STUDIES GROUP JOURNAL NO 28: 2014-1538

CSG Annual Conference - Belfast - April 2014 - Dundrum Castle

Fig. 1. Dundrum Castle from the south. From Francis Grose’s ‘Antiquities of Ireland’, Vol. 1,1791. The remnants of crenellations are seen on the round tower, with a gaping hole to the south.Today the vaulted roof section over this gap has also gone. The twin-towered? Gatehouse, withsmall rectangular windows to the rear is without its D-shaped fronts, as today. There is around-arched light in the Upper Ward outer wall in front of the round tower that does not existtoday - this part of the wall is reduced to its footings. Uncertain date to gate in the Lower Ward.

THE CASTLE STUDIES GROUP JOURNAL NO 28: 2014-15 39

Dundrum. Figs. 2, 3. Two views of the castle by John Bulman. c. 1800. © British Library . Ref:Shelfmark: Ktop LII Item number: 48.a. The heavily cropped lower view is probably from the west.Nassau Williams Senior, a 19th century economist wrote the following lines about Dundrum

Castle in his ‘Journal of a visitto Ireland, 1862’,‘The most in-teresting object is Dundrum cas-tle, finely situated on a hillabove the little town. It was builtby the Knight’s Templars. Ex-tensive outbuildings surroundan inner court, containing per-haps a couple of acres. In themiddle of this court, unconnect-ed with any other building risesa solitary round tower, aboutfifty feet in diameter, and sixtyor seventy feet high.’ The castlewas built by John de Courcy,following his invasion of Ulsterin 1177. It was built to controlaccess to Lecale from the westand south and would have origi-nally been built of earth andtimber. In 1204 de Courcy wasexpelled from Ulster by Hugh deLacy who strengthened the cas-tle with a massive round keep’.




ABOVE: Fig. 4. PlanBELOW: Fig. 5. Speculativeview of Dundrum from thesouth-east. From J. J. Phillips,‘Early military architecture inIreland’, 1883-84.



Dundrum Castle, County DownAn overview and excavation report by PhilipMcDonald, Excavation Director, Queen’sUniversity, BelfastThe castle, which is situated on the summit ofa prominent hill overlooking the small coastaltown of Dundrum and the adjacent tidal inletof Inner Dundrum Bay, has a long and com-plex history. The surviving upstanding re-mains consist of an inner and an outer ward.The inner ward dates to the Anglo-Normanperiod and contains the remains of a largecircular tower and an apparently asymmetricalgatehouse with a single projecting semi-circu-lar tower, whilst a seventeenth-century domes-tic structure, known as Blundell’s House, islocated in the outer ward. Artefactual andplace-name evidence suggests that the Anglo-Norman castle was, almost certainly, built up-on the site of an enclosed, high-status, Early

Christian settlement. Following the decline ofthe Anglo-Norman Earldom of Ulster in thefourteenth century, the castle was occupied bya branch of the Magennis family, althoughoccasionally it was temporarily possessed byother prominent Gaelic figures and variousrepresentatives of the Crown. During this later-medieval phase the upper floor of the circulartower within the inner ward was significantlyaltered. It is uncertain whether the outer warddates to the Anglo-Norman period or the laterGaelic phase of medieval occupation. Follow-ing the Nine Years’ War (1594-1603) the cas-tle and its associated estate was held byEdward Cromwell, whose son subsequentlysold it to the Blundell family. After beinggarrisoned during the wars of the 1640s, localtradition maintains the castle was slighted byParliamentarian forces in the mid seventeenthcentury. There is no evidence to suggest thatthe castle was occupied after this date, al-

Fig. 6. Dundrum Castle. View from the south. Illustration by Philip Armstrong. See PhilipArmstrong’s ‘Paint the Past’ website: http://philarm.com/ptpdefensive.aspx?PageSelect =defen-sive. Image © Northern Ireland Environment Agency. Note the approach to the Inner Ward.


Fig. 7. Dundrum Castle. Detail, highlighting the curtain wall with hourding, the round tower ofc. 1200 and the asymmetrical gatehouse (currently under review). A growing body of evidence issuggesting there was an original twin-towered gatehouse leading to a bridge over the rock-cut ditch.

though its associated manorial estate contin-ued to be farmed by tenants of the Blundellfamily. During the late eighteenth century thecastle and its estate passed into the hands ofthe Hill family, who carried out a considerableamount of landscaping on the castle to renderthe ruins more picturesque. Dundrum Castlewas placed into State Care by the Marquis ofDownshire in 1954.

Significant archaeological investigation of thecastle did not begin until the 1950s when aseries of excavations were undertaken at thesite by Dudley Waterman for the Archaeolog-ical Survey of Northern Ireland. On the basisof both the results of Waterman’s excavationsand architectural analogy, a consensus opinionemerged that the main phases of the buildingsequence at Dundrum were an earthwork bank

that possibly represented a temporary Anglo-Norman ringwork defence, followed by thepresent curtain wall of the inner ward, thecircular great tower, the inner gatehouse, thecurtain wall of the outer ward and finally Blun-dell’s House. Since 2009, three seasons ofexcavation at the site have been directed byPhilip Macdonald (Queen’s University Bel-fast) and Liam McQuillan (Northem IrelandEnvironment Agency). Although intended toinform the future management strategy for themonument, the excavations have been setwithin a research framework intended to ad-dress some of the unresolved historical ques-tions concerning the site.

The recent excavations have uncovered theremains of a near-circular, clay-bonded stonerevetment on the original summit of the hill



Fig. 8. Dundrum Castle. Aerial view - from the north - taken during the Time Team dig, in 2011(televised in June 2012 (Series 20 Episode 9).

which, in advance of radiocarbon dating, isassumed to represent a pre-Norman buildingplatform. The results of a cutting across the onesection of the defences to the west of the innerward where they consist of two, rather thanone, rock-cut ditches, suggests that the outerditch is all that remains of the pre-Normanenclosure. Re-excavation of one of Water-man’s trenches has demonstrated that theearthwork bank, which he identified as an An-glo-Norman ringwork dating to the campaign-ing phase of John de Courcy’s conquest ofUlster, is actually the remains of a series oflater levelling deposits that post-date the con-struction of the inner ward’s curtain wall, butapparently pre-date the construction of thegreat circular tower. Within the outer ward

excavation has uncovered the remains of amedieval lime kiln dating to the second half ofthe thirteenth century, built into a quarry edgeover five metres deep. Investigations withinBlundell’s House and in the extra-mural areasof the site have uncovered evidence for boththe construction of garden terraces specula-tively dated to the first half of the seventeenthcentury, and the transformation of the slight-ed castle and landscaping of its immediateenvirons in order to create a ‘picturesque’ruin during the closing decades of the eight-eenth century.

We hope to address a number of unresolvedresearch questions in future seasons of excava-tion, including the character of the gatehouseinto the inner ward. Waterman, basing his



opinion on unconvincing excavation evidence,thought that the present plan of the gatehousewith a single tower was the original one, butthis has long been considered problematic,especially as a c.1800 Downshire Estate Mapshows the gatehouse as having two projectingsemi-circular towers and Phillips’ 1883 plan ofthe Castle depicts the remnant stubs of the‘missing’ tower. Despite his awareness of theplans showing the gatehouse had two project-ing towers, Waterman based his interpretationon the assumption that the only practical lineof approach to the gate tower was from thewest through the area shown in the earlierplans as having contained the ‘missing’ tower.Re-excavation of Waterman’s trenches in thegatehouse will test the validity of his assertionthat no evidence for the ‘missing’ tower waspresent. If the double projecting tower plan forthe gatehouse is accepted as genuine thiswould indicate that the inner ward was origi-nally approached directly from the south, rath-

er than the west as at present (and as suggestedby Waterman). Such an approach would re-quire the ground level within the outer ward tobe considerably higher than it is today. It isprovisionally suggested that the rock-cut scarpslope in front of the gatehouse represents themodified remains of the inner edge of therock-cut ditch that originally surrounded theinner ward (and still survives to the west andnorth of the inner ward) and that the outer edgeof this ditch would have been defined by a,now removed, counterscarp bank, the base ofthe ditch being broadly equivalent to the lineof the modern path that provides access to theinner ward via its original entrance. This beingthe case, access through the gatehouse wouldoriginally have been provided by a bridgecrossing this outer ditch. It can further bespeculated that this counterscarp bank wasremoved when the outer ward was created, thedisplaced earth being used to fill the recentlyexcavated lime kiln and the associated quarry

Fig. 9. Ground plan of Dundrum Castle and its immediate environs produced by James Phillips(1883, pl. opp. 17; 1883-84, pl. opp. 160). Phillips clearly shows the stubs of the footings of thewest gatehouse tower forming the curve of the ‘D’.



Fig. 10. Details of various Ordnance Survey 6” and 1:10000 surveys and revisions of DundrumCastle and its environs. Top left;1: 1859. Top right: 2. 1903 only shows one stub; bottom left; 3.1925; and bottom right 1931. Only the 1925 version (3) vaguely shows two (presumptive?) Sets of‘D’-shaped footings. The 1903 map (2)1:2500 Co. Down sheet 44-5 is courtesy of Rodney Leary.

located in the south-eastern corner of the outerward. This would have necessitated the origi-nal entrance to the inner ward being reused asa means of providing access between the twowards. It is possible that the rock-cut scarpslope that formed the back of the ditch sur-rounding the southern edge of the inner wardwas further cut back to undermine the gate-house towers during the late eighteenth or earlynineteenth centuries as part of the efforts of theHill family to create a ‘picturesque’ ruin of themonument, resulting in the prominent over-hang noted by Waterman (and comparativelyrecently filled in with a mound of earth by thethen Environment and Heritage Service).

A second historical ‘problem’ is the location ofthe lesser tower recorded as being present atDundrum in the 1211-12 Pipe Roll, one of thefew extant historical references to the Anglo-

Norman Castle. One clue to its location is thepresence of a possible latrine chute in thenorth-eastern part of the outer ward’s curtainwall immediately adjacent to the gap in theinner ward’s curtain wall which is interpretedas representing the original entrance to thecastle. lf correctly identified, the presence ofthe latrine chute suggests that in the north-eastern part of the inner ward there was astructure - possibly the lesser tower attested inthe Pipe Roll which, given its location, mayhave also formed the original entrance into thecastle. It is intended to excavate this area of theouter ward in order to recover evidence of boththis structure and the character of the originaldefensive ditch hypothesised above as extend-ing across the southern edge of the inner ward.

Philip McDonald


1

34

2


Fig. 11. Dundrum Castle. The early C13 round tower in the Inner Ward with original first floorentrance. View from the south-east.


The Round Tower c. 1200-1210?John de Courcy’s contribution to the castle wasthe curtain wall and the rock cut ditch aroundit. In his time there were probably only woodenbuildings inside it. When Hugh de Lacy be-came Earl of Ulster in 1205, fortifications wereimproved in the earldom and he was probably

responsible for the construction of the, nowroofless, round tower. The main entrance wasoriginally at first-floor level for security (nowa gap on the SE side - fig. 11). The presentlower entrance was inserted in the 15th centuryand high above two stones remain from a boxmachicolation (fig. 13). The tower now has a


Fig. 12. Dundrum. Plans and section. Taken from the on-site display panel, after E M Jope, 1966.Jope and other writers consider the vaulted concentric chambers to be C15.


ground floor room with two chambers above,but the original arrangements of the upper part,especially the second (top) floor, are uncertainbecause of 15th century remodelling. Thoughit is a shell, the level of the floors can be tracedfrom offsets and joist holes (fig. 14).

Ground Floor: Internally it is 46 ft in diameterwith massive walls strengthened externally by abase batter and founded on solid rock. There are

two deeply-splayed narrow loops (now blocked)at this level, with arches built on plank centering;the marks in the mortar are clear. From this level,excavation has shown that a large pit was cut 23ft deep into the rock for water storage, fed byseepage. Wells are common in castle keeps butcisterns of this type are unusual. Originally, thisfloor would have been only accessible from thefirst floor above via the spiral stair.


Fig. 13. Dundrum. The round tower (c. 1210), from the east. Ground floor entrance, probablybroken through the wall in the C15. Staircase loops above. Two remaining corbels of the boxmachicolation added at roof level. Spiral stair immediately adjacent to the C15 entrance (left).


First floor: Now reached from the groundfloor by a typical Norman vaulted spiral stairin the wall thickness, which continues on up tothe roof level. The floor was supported by aninternal offset. This was once a grand andcomfortable room with a large hooded fire-place (now ruined) and two windows withseats. At this level there were two originalentrances, the principal entrance from the wardon the south and a secondary narrow access

door to the east, that presumably led to thecurtain wall walk via a short wooden bridge.Two garderobes are situated close by on thecurtain. There are no garderobes in the tower.

Second Floor: Before the 15th century chang-es this would have been another grand room,but more private at this level. Some of thepresent arrangements around the perimeterseemingly belong to the C15. The floor level ismarked by joist holes. In the wall thickness


Fig. 14. Dundrum. Interior of the round tower looking west. Showing (1) Offset and joist holes for 1stfloor. (2). Narrow door that gave access to the wall-walk and latrines via a timber bridge. (3). Joistholes for the 2nd (top) floor. (4). One of the 4 doors (or open apertures) giving access to the concentricmural chambers and allowing the ingress of borrowed light. It is uncertain whether these latterfeatures, and the interconnecting perimeter, concentric rooms are original features (see fig. 15). Theroof to these chambers is a mixture of wicker-work centering and bird’s nest/interleaved corbelling.

there are a series of what can described assmall interconnecting chambers which allopened into the main room. These perimeterchambers are roofed either with overlappingflags (corbelling - C on the plan, fig. 12) orvaults showing the impressions of wickercentering (W on the plan), quite different fromthe plank centering at the lower levels. Thewall-walk and parapet, once important defen-sive features are largely ruined but the numer-

ous holes for rainwater drainage at the base ofthe parapet are clear. Round towers are notcommon in Ireland and in Britain are best repre-sented in South Wales as at Pembroke. Masonsfrom Pembroke were prominent in late-12thcentury Ulster and Hugh de Lacy had strongWelsh connections. This could be the back-ground for the Dundrum tower, a more up-to-date, fashionable design than the rectangulartower at Carrickfergus. There are similarities to


1

2

3

4


the Pembroke round tower. Although on a muchsmaller scale, the structure is conceived andconstructed to the same formula. A vaulted spi-ral stair within the wall thickness from basementto wall-walk. No garderobes with the tower, buta small passageway leading to a doorway givinga bridge access to the latrines upon the nearbywall-walk. A timber constructed entrance lobbycantilevered at first floor. Indications are that thebuilding post-dates Pembroke and was probablycompleted by 1210. Further discussion of thesecond floor concentric gallery is found in thefollowing figure captions (figs. 16-17).


Fig. 15. Dundrum Castle. The top (3rd) storey showing 3 of the probably 4 doors or aperturesgiving access to the vaulted mural concentric chambers with the wall-walk located above thevaulted roof and arranged with high volume series of rainwater drain holes. Waterman & Jopedated the mural chambers to the 15th century because one has a wicker-centred vault, which wasthen thought to be diagnostically of that date. Since then wicker-centred vaults have been foundin 13th century contexts: secular (Castle Roche, Co. Louth) and religious (Abbeyknockmoy, Co.Galway). It is perhaps easier to see the vault at Dundrum as another early use of the techniquethan to argue that the upper wall was partially rebuilt later.

Further Reading - Dundrum CastlePhillips, J. J., ‘Military defences of the Anglo-Norman donjon at Dundrum, Co. Down’, inThe Irish builder, Vol. XXV, No. 561, May 1,1883, pp. 144-145.Waterman, D. M., ‘Excavations at DundrumCastle’, in Ulster Journal of Archaeology(third ser.) 14, 1950, pp. 15-19.

Waterman, D. M., ‘The water supply at Dun-drum castle’, Ulster Journal of Archaeology(third ser.) 27, 1964, pp. 133-5.Jope, E. M., (ed.) 1966. An ArchaeologicalSurvey of Co. Down, pp. 207-211.Hamlin, A., Dundrum Castle, Co. Down., Bel-fast, HMSO), 1977.McNeill, T., ‘Squaring Circles: flooring roundtowers in Wales and Ireland’, in The MedievalCastle in Ireland and Wales, 2003, 97-99, 102-6.Sweetman, D.,The Medieval Castles of Ire-land, 2000, pp. 36-7, 58-9, 86-88, 105.Macdonald, P., 2011. Excavation at Blundell’sHouse, Dundrum Castle, County Down, 2009,(Centre for Archaeological Fieldwork DataStructure Report. No. 71). Centre for Archaeo-logical Fieldwork, Queen’s University Belfast,Belfast.(http://www.qub.ac.uk/schools/CentreforArchaeologicalFieldworkCAF/PDFFileStore/Filetoupload,236105,en.pdi).



Dundrum: ABOVE. Fig. 16. Detail of fig. 9. Clearly illustrating the putative stubs of the ‘D’ shapedfrontage to the west gate-tower. For the Jope plan see Sweetman, 2000, p. 58. BELOW: Fig. 17.The remains of the gatehouse from the south (Barbican) Lower Ward, which was heavily land-scaped in the late C18 and early C19.


Fig. 18. ABOVE: Blundell’s C17 house in the Lower Ward. BELOW: Fig. 19. The entrance arch intothe lower ward, as depicted by Grose (fig. 1). Date of this arch is, as yet, undetermined, but isdepicted in the illustrative recreation (fig. 7).


Documents

Fig. 1. Dundrum Castle from the south. From Francis Grose ... · 38 THE CASTLE STUDIES GROUP JOURNAL NO 28: 2014-15 CSG Annual Conference - Belfast - April 2014 - Dundrum Castle Fig