The house that Ludwig built

After he had published the Tractatus Logico-Philosophicus in 1922 Wittgenstein, believing that he had finished with philosophy, took up school teaching. His intolerant temperament made him wholly unsuitable to the occupation and it is not irrelevant to the theme of this book that the occasion that led to his giving up teaching was his anger at a child's failure to draw Corinthian columns correctly.

His sister Margaret, in her anxiety to find a "normal" job for him, invited him to join with her and her architect in the design of a large mansion in Vienna. Wittgenstein, a friend of the great architect Adolf Loos, had at one time seriously thought of becoming an architect and so in early 1926 he entered the fray with an all-consuming commitment.

The architect that Margaret had chosen for her house was the man whom Loos called his "most loved student'', Paul Engelmann, a close confidante of Wittgenstein during the Tractatus years.

This book is the account of the development of the design and the gradual assumption by Wittgenstein of the primary role of architect which led Engelmann graciously to withdraw from that role, and from that friendship. What is remarkably important about the book is the patient chronicle of the evolution of that design through ten stages (including a complete change of site). These are illustrated fully for the first time from Engelmann's sketch-book and for this alone we should be very grateful. Also revealed is the later participation in the process of production drawings and site supervision of another of Loos's pupils, Jacques Groag. This is an important extension to the familiar story. There are no authenticated drawings by Wittgenstein and neither he nor Engelmann was capable of preparing the necessary technical information. Groag had been Loos's assistant on the celebrated Moller house in Vienna and the account of his contribution therefore fills a hitherto mysterious vacuum in the story.

The book raises three questions. What is the relation between Wittgenstein's philosophy and architecture? How does he stand in the company of other dilettante-architects? What are the intrinsic qualities of the Kundmangasse house?

A number of remarks on architecture in the writings of Wittgenstein point to a view that conforms closely to the celebrated statement of Adolf Loos: "The work of art is brought into the world without there being any need for it. The house on the other hand satisfies a need . . . Only a very small part of architecture belongs to art: the Tomb and the Monument. The rest, everything that serves an end, should be excluded from the realm of art.'' Thus, for instance, Wittgenstein writes: "Architecture immortalises and glorifies something. Hence there can be no architecture where there is nothing to glorify"; and "architecture is a gesture. Not every purposive movement of the human body is a gesture. And no more is every building designed for a purpose architecture.'' These remarks confirm the dichotomy "architecture vs. building'' that stemmed from the application to architecture of Kant's definition of aesthetics and the work of art as "purposiveness without purpose''. Such an application is completely at variance with the classical concept of architecture as deeply purposeful, a Practical Art "that serves an end other than itself"; and Kant himself explicitly stated that the formula of "purposelessness'' could not apply to architecture which could never be "an end in itself''. But the travesty took hold; architecture as "frozen music'' provoked the counterblast of "functionalism''. At least it would appear that in Wittgenstein's mind at the time a house was not a proper subject for "architecture''.

If on the other hand one applies to architecture Wittgenstein's later thinking that centres on the concept that "the meaning lies in the use'' we arrive at a very different conclusion. Within the truly classical (as opposed to "Classicist") concept of a Practical Art the concepts of the useful and the beautiful were not in opposition but were bound together in a single word kalon: and Wittgenstein's later philosophy introduces us to such a rich interpretation of the modes of use that we begin to see that the disjunction between "that which serves an end'' and "that which is an end in itself'' is false to the nature of architecture. In his lecture on ethics of 1929 he wrote that "In order to get clear about aesthetic words you have to describe ways of living . . . connected with all sorts of other gestures and actions and a whole situation and a culture''. This is far from the dismissal of need from the realm of formal significance.

Wittgenstein's status as "dilettante'' (Wijdeveld's word) architect is unusual. There is a time for the contribution of the amateur, a time when an established set of rules is widely acknowledged and embodied in a range of familiar models, when there prevails a technology based on comprehensible methods and executed by accessible craftsmen, and when the requirements for the building in question were open to common consent. (Thomas Jefferson's campus for the University of Virginia and his Monticello are cases in point.) But for Wittgenstein none of these pre-conditions existed. There was no accepted building language, or technology, or model, and no agreement about the nature of a new palais. In this sense his contribution was remarkably original. His own comment was that it was an exercise in good taste by which he meant a restraint from indulging in unwarranted "gesture"; and certainly it resisted the Viennese penchant for indulgent decoration. (His father had paid for Olbrich's Sezession pavilion.) But in practice it went beyond mere passivity and one is tempted to suggest that it anticipated the claim by Mies van der Rohe that "less is more''.

As to the qualities of the building that emerged, we can immediately dismiss a few favoured assumptions. It has nothing to do with the raumplanung of interlocking spatial volumes that was Loos's priority. Nor did it reflect any of the avant-garde developments elsewhere in Europe, least of all that of Le Corbusier whose famous "Five Points of Architecture'' were published at the same time and whose Villa Savoye was exactly contemporary with the Wittgenstein house. (It is typical of the perennial vagueness on the subject of architecture in English literary circles that Keynes should refer to the house as "a la Corbusier".) It did not engage in the current discourse but pursued its own inner logic solipsistically.

Its one crowning achievement centres upon the entrance hall. In the first place this is remarkable as a recall and comment upon the home of Wittgenstein's childhood. It is a condensed abstraction of the entrance hall, honorific staircase and focal sculpture of the Wittgenstein family palais in the Alleegasse, to which it stands in relationship as a chamber work by von Webern to a processional march by Handel. This design episode first appears in Engelmann's fifth scheme and, once there, freezes out any further attempts by him to adopt an irregular Loosian disposition of rooms: and one may well suspect that it was a response to the first moment of intervention by Wittgenstein (in collusion with his sister) when he joined in the "collaboration'' with Engelmann. But it is above all an area within which all that Wittgenstein had to say could be shown with the greatest clarity. Furthermore it could retain that character "untamed'', as he might have said, by the soft furnishings, objets d'art and Viennese "good taste" of his sister. It is a haunting space recalling the films of Andrei Tarkovsky and the world of Giacometti's tall goddess figures. Its greatest celebration is the sequence of black and white photographs in the 1972 documentation by Bernard Leitner that helped to save the building from demolition.

Wijdeveld argues that Wittgenstein "stabilised'' Engelmann's facade design by adjustments to the intervals between door and window elements and the periphery of their planar ground. This appears to be so: but there are no grounds to suppose that Wittgenstein adopted a proportional system in the Albertian sense of a common ratio relating part to part as part to whole.

Where proportion (not proportional system) becomes an original and forceful element itself is in the overall form and subdivision of the glazed doors and windows. The originality and refinement of detail here is extraordinary: screwed steel angle sections of special profile receive the glass panes and hardwood beads: and every catch, lock and door handle were designed de novo by Wittgenstein in negotiation with anguished sub-contractors. What is awesome is the height of the doors (up to 3.10m/10'4'' high) with glass panes whose vertical ratio is 10:1.) These dimensions were possibly a transfer to the Kundmangasse of Margaret's previous house, the "Palais" Batthyany-Schonborn by Fischer von Erlach, but in the context of otherwise quite modest dimensions they have an unforgettable power.

The book is a large and handsome production with excellent plates including beautiful drawings by Hermine Wittgenstein which are the only record of the way in which the interiors were furnished on completion in 1928. One of the photographs is a telling indictment of the recent "conversion'' of the building. It shows the replacement of the wall and connecting door between library and reception room by a brash sliding metal shutter, which is a parody of Wittgenstein's minimalist devices.

Wijedeveld understandably confines his theoretical discussion to the thinking and terms derived from the Tractatus. But insofar as Wittgenstein remains a source of inspiration and instruction for artists and architects it is paradoxically his later philosophy exploring the breadth and uniqueness of "forms of living'' by interpretations of "the meaning that lies in the use'' that offers quite a different, life-enhancing programme for architecture.

As for the house itself, it remains a mystery. In its day and for long after, it passed unnoticed. But ever since it was "discovered" in 1970 and published in the avant-garde magazine Art forum it has been presented as a precursor of the mode of design known as Minimalism. However, it will never fit into such categorisation but will remain in the history of 20th-century architecture as a "rogue" and an enigma.

Colin St John Wilson, formerly professor of architecture at the University of Cambridge, is currently architect of the British Library, St Pancras. His house and studio in Cambridge, designed by himself, now houses the Wittgenstein Archive.