Precision in Architecture

Certainty, Ambiguity and Deviation

Mhairi McVicar

• Éditeur: Routledge
• Année: 2019
• ISBN:
  9781351838504
• Édition: 1
• Sujet: Architecture
• Sous-sujet: Architecture générale

This book offers a detailed insight into the desire for, and consequences of, precise communications in the daily life of contemporary architectural practice through close readings of constructed architectural details by Sigurd Lewerentz, Caruso St John Architects, Mies van der Rohe and OMA.

In the professionalised context of the contemporary architectural profession, precise communications – drawings, specifications, letters, faxes and emails – are charged with the complex task of translating architectural intent into a neutral and quantifiable language which is expected to guarantee an exact match between the architects' intentions and the constructed result. Yet, as any architectural practitioner will know, it is doubtful whether the construction of any architectural project may ever exactly match all written and drawn predictions. This book challenges claims to certainty which have been attributed to such communications from the mid-nineteenth century onwards, and critiques ongoing expectations of certainty in contemporary architectural production.

Part I

Two projects

FIGURE 1.1 Constructing a flagstone and concrete block wall.

1

A precisely ambiguous wall

The wall, as constructed, is 519.5mm thick, plus or minus the irregularities of flagstone. It consists of: 12.5mm plasterboard, a vapour barrier, 145mm timber studs, 150mm ‘Rockwool’ insulation, 12mm far-eastern marine-grade plywood, ‘Tyvek^®’ building membrane, a 50mm cavity, stainless-steel wall ties, an unforeseen layer of 100mm concrete block and a pragmatically redundant skin of flagstone, varying in width from 200–300mm and recycled from the 1840s longhouse upon whose footprint this self-build residential project was constructed.

This wall was the result of five years of planning, sketching, drawing, specifying, sourcing and self-building, and more than a decade of architectural education and architectural practice in two countries. It was planned meticulously. Computer-aided design (CAD) drawings specified the exact number of concrete blocks and timber studs to be laid out according to modular dimensional systems using standard timber sections, plywood panels and plasterboard sheets. Off-the-shelf components, selected for logistical and economic efficiencies and the ease of self-build construction, supported an architectural intent of neatly, simply, inserting a new timber-frame dwelling within the ruins of an 1851 Orkney flagstone longhouse. Acting as architect, client, project manager and contractor, this small self-build project should have been straightforward for us to construct. Our team of three anticipated complete control over an exact alignment between our conceptual intent and the constructed result.

Instead, of course, the project is constructed with adaptations, uncertainties and ambiguities. The wall, as constructed, is conceptually and physically ambiguous. It deviates from the original intent, remaining in place as a gap between the idealised and the actual.

Laden with the weight of cultural, archaeological, geological and historical referencing, as well as internalised architectural ideologies borne from an architectural education under a Miesian curriculum, this seemingly simple wall physically embodied a sense of uncertainty I felt as an architect. The construction of this wall raised a question about recommendations, repeatedly made to the architectural profession, that precise instructions made in advance of construction should promise certainty, and, in turn, assure quality.

The one certain opportunity

‘The one certain opportunity’, Francis Hall wrote in the Technical and Practice section of the Architects’ Journal in 1994:

available to an architect to set down a definitive and enforceable expression of standard and quality is by way of a properly drafted specification. If this is done, there is understanding and certainty all round. If it is not, there is often disagreement and disappointment.¹

‘The objective,’ Hall emphasises, ‘must be certainty’.²

This recommendation reflects a long-held belief. In 1812, a Parliamentary Select Committee critiqued architect James Wyatt for ‘loose and inaccurate’ instructions at Somerset House and the Houses of Parliament and recommended that ‘precise arrangements should be made’ to guarantee the cost and quality of publicly funded works.³ A Select Committee in 1828 similarly sought ‘precise specification and careful superintendence’ and the explicit avoidance of ‘all deviations from the original plan.’⁴ Precise instructions in advance of construction are today still charged with guaranteeing certainty in any translation of an architectural intent into a constructed reality, as a more recent promise attributed to Building Information Modelling (BIM) promises:

It follows that the absence of conflict in the information given to the Contractors means that there should be no problems to resolve during construction.⁵

The specified aim of ‘no problems to resolve during construction’ will be familiar to many practising architects. The objective, it is recommended, must be certainty.

FIGURE 1.2 The project, as proposed: perspective.

FIGURE 1.3 The project, as proposed: axonometric.

The wall, as proposed

The pursuit of certainty in our self-build in the Orkney Islands in northern Scotland had started simply enough. We had purchased the ruins of an 1851 flagstone longhouse and had paced off a plot of land for exchange with the farmer who owned the land. Sitting on a narrow stretch of land between Atlantic cliffs to the west and a wide sandy bay to the east, the remaining walls and collapsed roof sat on an east–west axis, the north side buried into the slight slope of the site, western gables thickened against prevailing westerly winds, windows facing south to capture the extremities of summer and winter solstice sun in a northern climate. The flagstone which made up the walls, roofing and flooring had been quarried from easily split layers from exposed beds on the coastline. An abundant material in an island landscape bereft of timber, these flagstones carried a geological, historical and cultural weight.

In Orkney, flagstone was used to construct Neolithic dwellings, burial chambers and stone circles. It was carved with decorative symbols by Neolithic and Viking populations and is captured in poetry describing its significance in Orkney as protective, enduring, symbolic. ‘Hearth stone, water niche, lintel’, twentieth-century Orcadian poet George Mackay Brown wrote of flagstone: ‘I can’t work wood’.⁶

The ruins of flagstone walls were a compelling starting point for our architectural intent. Trusting to the intuition and embedded knowledge of the nineteenth-century self-builders who had sited the longhouse in this exposed landscape, we proposed to retain the poetic, cultural, tectonic and protective flagstone shell. We planned to insert an insulated, dry, clean and easily assembled timber frame into the shell of the flagstone. This conceptual intent was to satisfy our amateur building skills, our need for contemporary comfort, our architectural idealisation of achieving a clearly defined, truthful, gap between the old and the new. As recently licensed architects we hoped, of course, to set out our architectural ideology with this small project. Concepts of truth and resonance lurked behind our insistence that the flagstone wall be retained. A gap between the ‘new’ and the ‘old’, as important conceptually as it was physically, offered tolerable contingency between the meticulously predictable standardisation of a contemporary timber frame and the unpredictability of an uneven, imprecisely hand-built flagstone wall, a condition which we struggled to capture with the detailed CAD drawings of our proposal.

The wall, as proposed, was to be 525.5mm thick, plus or minus assumed tolerances for the existing flagstone. It was to consist of: 12.5mm plasterboard, 12mm plywood, 150mm timber studs, 150mm sheep’s wool insulation, ‘Tyvek^®’ building membrane, a 50mm cavity and existing 300–400mm flagstone walls, as the existing condition to which the precisely specified and drawn timber kit would be adapted. The concept was clear, predictable and meticulously drawn and costed, accommodating our predictions of contingency, risk and construction tolerance in both the preparation and assembly of materials. We were certain of our intent and of its translation into construction.

Of course, our project did not go as planned. Fifty years of exposure without a roof or collar ties had left the wall more unstable than first predicted by an initial site visit. Our drawings had predicted that we would demolish unstable areas of the wall, and insert a stabilising timber frame in the remaining structure. Instead, a recommendation from a locally experienced building inspector directed us towards the full demolition of the flagstone wall, rejecting the existing wall as unstable, unsalvageable. This was not an option we had previously considered. The removal of the existing wall was a significant unplanned deviation from our otherwise meticulously planned set of construction documents. More importantly, it fundamentally challenged the core architectural intent of our project, that of the bringing together of the poetic, historic, contextual in-situ flagstone wall with the precise order of a contemporary modular construction. To lose the flagstone wall would be to lose the architectural intent of the project.

Faced with this uncertainty, we sought to re-establish the certainty of our architectural intent. Perhaps, we argued to ourselves, a rebuilt flagstone wall could be in itself interpreted as truthful as a demonstration of the inevitability of deviation encountered in almost any architectural project. Reserving our guilt, we dismantled the flagstone wall and restacked the stones nearby in preparation for reuse. The timber frame was quickly completed on the grounds of the footprint of the removed wall by the end of a summer of construction, clad in plywood sheathing, and wrapped in a weatherproof membrane. We now faced the potential of leaving the membrane exposed over winter. Advised against this by the membrane manufacturer, as well as by our own growing knowledge of the site’s extreme exposure, we were again self-dir...