Unsupported browser

For a better experience please update your browser to its latest version.

Your browser appears to have cookies disabled. For the best experience of this website, please enable cookies in your browser

We'll assume we have your consent to use cookies, for example so you won't need to log in each time you visit our site.
Learn more


  • Comment

On 3 August 1910, Edwin Lutyens wrote to his wife: 'I do wish he did not want a castle but a delicious lovely house with plenty of good rooms in it.' But the client, Julius Drewe, was not to be persuaded. The founder of the Home and Colonial Stores and a self-made millionaire, Drewe had retired at the age of 33 in order to establish himself as a country gentleman. His choice of location, a granite outcrop 270m above the Teign Gorge on the edge of Dartmoor, was determined by his claim to be descended from the land's one-time owner, a Norman baron named Drogo de Teigne.

Only a castle would do.

By 1911 the Drewes had approved a scheme which drew on elements of Lutyens' earlier houses - the U-shaped plan, open to the north, was reminiscent of Papillon in Leicestershire and Goddards at Abinger in Surrey - but had the scale (and the gatehouse) of a full-blown castle.

The following year they had a change of heart and decreed that the plans were too generous for their needs. However, work had already begun. The foundation stone had been laid on Drewe's 55th birthday in 1911, and so the design was truncated rather than scaled down.

This pared-down scheme comprised the east wing and the splayed service wing to its north, along with the foundations of the south range. The principal living rooms, originally intended to occupy the south range, were transferred to the east wing; the basement of the south wing was redesigned as the chapel; and the west wing was dispensed with altogether.

It was still a monumental undertaking, and one which was not completed until 1930, a year before Drewe's death.

The intention was to provide 20th-century comfort within a fortress of solid granite walls; the oors and the at roof were constructed of concrete and steel. Lutyens' solution to the issue of water penetration was simple: asphalt, a then new material brought from pits in the Caribbean. The 60-90cm solid walls were built with a vertical layer of asphalt between internal and external faces. This vertical asphalt skin was carried through to roof level, where it joined a horizontal asphalt screed.

Mehmet Berker of Inskip + Jenkins, project architect for the current programme of repairs at Castle Drogo, says: 'Lutyens did not really understand how asphalt works. He thought of it as a panacea curing all ills. He assumed that it could act like a dampproof course, tanking and waterproofing, so he used it on the horizontal, on the vertical and on the horizontal again. But of course it gets restrained and it cracks. It was leaking from day one.'

The problems were - and are - compounded both by the exposed, wet site and the extreme austerity of Lutyens' design.

Exterior detail is, to quote Pevsner, 'minimal Tudor: mullioned and transomed windows, without dripstones, set in vast expanses of sheer ashlar walling of local granite, carried up without interruption to an irregular crenellated parapet'. Only the entrance is less austere, with moulded plinths to the polygonal corner turrets, and a corbelled-out central oriel. The absence of waterspouts or dripstones makes the entire facade susceptible to rain.

There is ample evidence of the Drewes' increasingly desperate, ad hoc attempts to assuage the invasion of water through the at roof, the huge leaded windows and the mortar joints in the granite walls. They seem to have borne the matter in surprisingly good humour, gamely lifting up the heavy paving slabs on the rooftop to chase away the water underneath and, touchingly, keeping the matter from their architect so as not to hurt his pride. But they were fighting a losing battle. After the Second World War the rooftop slabs were removed altogether and put into storage, though nobody can quite remember where.

In 1974, fearful that the newly elected Labour government would introduce a wealth tax, the remaining Drewes made Drogo the first 20th-century property to be gifted to the National Trust.

Announcing the acquisition (AJ 02.04.75), a youthful Gavin Stamp felt able to reassure readers that 'the only important work necessary was to repair the at roofs and insert a new damp course below the parapet'. By 1979, the AJ was running a news story with the headline 'Leaking Lutyens to cost a fortune' (AJ 25.07.79) as the true scale of the problems began to emerge.

The trust launched successive attempts to make the castle watertight, but to little avail. Between 1983 and 1989 the parapets were dismantled to enable a new lead damp course to be inserted, preventing water ingress from the top of the building, and new asphalt was applied. Although this was successful in the short term, it wasn't long before the asphalt began to crack. As Peter Inskip of Inskip + Jenkins says: 'A lot of architects have spent a lot of time trying to solve the damp problem but it became blatantly obvious that no one had actually investigated the real cause.'

It fell to Inskip + Jenkins to carry out a comprehensive investigation into the underlying causes of the defects. Although phase one of the current repair programme started on site in June 2006, Inskip is quick to add that 'it took us a decade to get to that point', making it one of the most intensive programmes of conservation and building research to be carried out by the trust.

The research emphasised the importance of using materials which would act as a waterproofing membrane and, crucially, would ex in response to environmental conditions.

The parapet walls, bell tower and three courses of stone from the bottom of the west wall have been removed to facilitate the installation of a Bauder waterproofing membrane and the insertion of Rubberoid damp courses over the parapet walls. An insulation layer will be laid over the membrane, then a drainage layer topped with fine gravel. The granite slabs, including the central dais, will be replaced according to Lutyens' original design.

The chapel wing is being repointed using a mortar that, while lime-based, will not leach 'free lime' that would stain the castle walls. Windows are being removed one by one to allow for the repair of the leading, and the replacement of the linseed-oil putty which seals the junctions between the windows and granite walls.

Drogo is, effectively, in a state of semi-undress. Sections of timber panelling have been removed to allow access to the walls behind. Swatches of blotting paper are dotted around the windows in an attempt to absorb and monitor water ingress. Where windows have been removed for repair, makeshift screens shield the openings, concealing outdoor views.

The castle bears the indignity rather well. The temporary absence of windows compounds the disorientating geometry, in a way which one imagines Lutyens might have rather enjoyed.

The two different ranges have different orientations and floor levels, allowing for spatially complex circulation space. It is easy to believe that you have burrowed deep within the castle when in reality you have returned to the building's edge, a reading which is encouraged by the presence of an internal window in the drawing room overlooking the internal stair. Doors in the temporary screens, designed to give builders access, allow the uninitiated to imagine that, rather than shielding the perimeter, the screens are a threshold to another world.

There is a Marie Celeste-like poetry to the great dustsheets draped over fittings and furniture marooned in the centre of the room; an incidental installation which suggests the seasonal flux of a country seat - the 'semi-slumber' which characterises the periods when the owner is not in residence and the grand public spaces become a stagnant top-water to the bustle of servant activity downstairs. Paradoxically, the overt impression of absence implies inhabitation far more intensely than the carefully conserved 'as found' interiors.

Similarly, there's a sense in which the areas in the drawing room where the timber panelling has been removed to reveal the solid construction behind, convey more about the architectural aspiration than the excellent, but rather literal, temporary exhibition on Castle Drogo's construction and conservation, which has been mounted in a replica of one of the original builders' huts.

It is hard to imagine a more pithy encapsulation of 'comfort within a fortress' than the juxtaposition of internal and external skin; the obvious signs of repair an evocative reminder of the difficulty of reconciling the two. Even the scaffolding, which creeps around the walls as the castle's windows are repaired, seems to underline rather than detract from Drogo's essential gravitas; a delicate tracery compared with the unadorned expanse of Lutyens' granite facade. For all its weightiness, there is an air of determined informality which is able to accommodate compromise.

Scaffolding and site huts, while clearly imposters, do not have the destabilising effect they would have had on a symmetrical form.

With characteristic eloquence, Pevsner remarked that 'the completed fragment has a compellingly abrupt sublimity that would have been diluted in the more ample and even arrangement of the earlier plan'. Its asymmetry bestows a sense of near-geological permanence, allowing the castle to be read as part and parcel of the rocky outcrop on which it stands; both powerful and picturesque. In retrospect, we can be grateful for the Drewes' retreat from their original plans, but also for their refusal to defer to their architect's advice. A 'lovely delicious house' may have kept the rain at bay, but it would scarcely have been a match for the forbidding drama of the site.

Credits Tender date 19 April 2006 Start on site 24 July 2006 Contract duration 31 weeks Gross external floor area Chapel: 230m 2; castle: 1,073m 2Total: 1,303m 2Form of contract JCT IBC 2005 Total cost £426,846.65 Client The National Trust, Devon and Cornwall Region Architect Peter Inskip + Peter Jenkins Architects: Peter Inskip, Stephen Gee, Mehmet Berker, Simon Perera, Maja Polenz Structural engineer Ralph Mills Associates Services engineer FHP Engineering Services Solutions Quantity surveyor Sawyer and Fisher Planning supervisor LHC Safety Management Main contractor St Blaise Selected subcontractors and suppliers Roofing Bauder; metal conservator Eura Conservation; paint analysis Catherine Hassall; sculpture Andy Mitchell; lime mortars Telling Lime, Arte Construco, St Astier; moisture monitoring Ridout Associates; cloaks Ruberoid Building Products; sealers Adshead Radcliffe; thermography and radar GBG; lead came supplier Kyme Studios; asphalt Clegg and Shortman; latex screed SBD; surveying and restoration George Bollard Geotechnics

  • Comment

Have your say

You must sign in to make a comment

Please remember that the submission of any material is governed by our Terms and Conditions and by submitting material you confirm your agreement to these Terms and Conditions.

Links may be included in your comments but HTML is not permitted.