In 1992 Michael Brawne published From Idea to Building, which he described as 'a critical view of the assumptions which influence initial design decisions and of the processes of development from inception to inhabited building, together with an analysis of the general implications of the design process'.
It was a learned, yet accessible, discourse on the subject of how architects turn programmes into formal solutions, of the kind that one would expect from a distinguished academic and practitioner of his generation (born 1925). He acknowledged the place of scientific method, the use of models (in the sense of successful precedents), and imaginative creativity. The design process retained its mystery, but it was possible to describe with a degree of rationalism the ways in which architects design.
Brawne died in 2003 and his last book, Architectural Thought, is now published posthumously. It revisits the question of how architects design - the subject of the previous book - but does not provide more detail or take the investigation into new territory. There is considerable overlap between the content of the two books, and this new one has a less consistent narrative than From Idea to Building.
Having said that, Architectural Thought is an intelligent, provoking and rewarding book in its own right. The structure is episodic, with chapters named 'Travel, books & memory', 'Thinking and drawing', 'Looking at pictures', and so on, which are almost self-contained essays. A lifetime's learning is worn lightly, and an intelligent first-year architecture student could follow the arguments and benefit from them. Brawne illustrates his consideration of the different theories of design by many examples, from a catholic range including Gehry, Schinkel, Libeskind, Kahn and Soane.
Brawne's preferred explanation of design, contained in both books, is the process usually identified by reference to Karl Popper, and expressed as P1 > TS > EE > P2, etc. The identification of the problem is followed by a tentative solution, which is drawn from known models. A critical examination against the programme leads to error elimination, which produces a revised solution that restates the problem in a modified form.
This cyclical motion continues until a satisfactory conclusion is reached.
What is pleasing about the author's description of this idea, and typifies most intelligent making of architecture, is not only the often-untidy combination of scientific method and pragmatism, but his liberal recognition that the process is never innocent; the designer always has choices, which are made with prejudice. Brawne is happy for history to be misinterpreted, as long as the reading leads to a critical starting point for something new. Similarly, he is relaxed about the designer's memory bank consisting largely of photographs of buildings, which convey only a partial version of their reality. If the process is useful, then the end justifies the means.
Within this general explanation of design, Brawne identifies two tendencies - continuity and innovation. The use of typology favours continuity, while functionalism leads to innovation. I think this is too simplistic, and exaggerates the opposition between the two. (He appears confused about typology anyway, unsure whether it is based upon use or upon form. ) BedZED, for example, surely expresses continuity in the use of the historical terrace type, while being functionally and technically innovative.
This is an elegant and useful book, written by a thoughtful and precise architect.
However, there are a few names spelt wrongly, which I suspect would not have got past Brawne's scrutiny.
Joe Holyoak is reader at Birmingham School of Architecture and Landscape, and an architect in private practice