Von Bart Lootsma.
Von Bart Lootsma.
It is in the fundamental works of philosophers like Michel Foucault, Félix Guattari and Gilles Deleuze that one should seek the theoretical reason for the use of diagrams in design work in architecture. In the first chapter of Capital and Schizophrenia,2 a text that architects have more or less forgotten or suppressed, Deleuze and Guattari write, ' Capital is indeed the body without organs of the capitalist, or rather of the capitalist being. But as such, it is not only the fluid and petrified substance of money, for it will give to the sterility of money the form whereby money produces money. It produces surplus value, just as the body without organs reproduces itself, puts forth shoots, and branches out to the farthest corners of the universe..' At the time (the year 1972), Deleuze and Guattari could not go too far. At most, they could sketch out a parallel between desiring production and social production: ‘…..the forms of social production, like those of desiring production, involve an unengendered nonproductive attitude, an element of antiproduction coupled with the process, a full body that functions as a socius. This socius may be the body of the earth, that of the tyrant, or capital. This is the body that Marx is referring to when he says that it is not the product of labor, but rather appears as its natural or divine presupposition. In fact, it does not restrict itself merely to opposing productive forces in and of themselves. It falls back on (il se rabat sur) all production, constituting a surface over which the forces and agents of production are distributed, thereby appropriating for itself all surplus production and arrogating to itself both the whole and parts of the process, which now seems to emanate from it as a quasi cause. Forces and agents come to represent a miraculous form of its own power: they appear to be “miraculated” (miraculés) by it. In a word, the socius as a full body forms a surface where all production is recorded, whereupon the entire process appears to emanate from this recording surface.’'3
Nowadays, to establish a relationship between desiring and social production no longer poses a problem: the computer takes care of it. As a sophisticated tool, the computer not only influences every particular aspect of social life, but above all it establishes a multitude of new relationships. Everything is indeed converted into a flow of data, an infinite interpolation of 0 and 1, which can be apparently exchanged and manipulated without difficulty. This flow of data has assumed the role of capital and even more than that. We live today in a 'space of flows', to borrow Manuel Castells' expression,4 within infinite networks, in which machines, men, desires and merchandise are linked to and converted into one another. It is a space that gives rise to altogether new relationships of power, symbolized by the new financial centres and cities that are springing up at exotic and improbable sites even today, Shenzhen, Singapore, Kuala Lumpur or Hani Rashid's virtual stockmarket. It is a space in which there is not even an immediate link between money and gold reserves. Everything simply merges in a series of metamorphoses whose beginning and end are impossible to situate. It is here that society produces it own delirium: 'Everywhere it is machines –real ones, not figurative ones: machines driving other machines, machines being driven by other machines, with all the necessary couplings and connections’.5 Nevertheless, this situation does not resemble the stammering and stuttering 'Society of the And' described by Roemer van Toorn.6 What energy has not been spent, over the years, so that these couplings and connections take place in fluidity and silence? Since the old telephone exchanges, with their plugs and loud switches, and Hollerith's crackling perforated card machines, to the supple software programmes of computers, immense progress has been accomplished. Nowadays everything has become process and the world is adapting. Data are indeed the bodies without organs of the second modernity.
For architecture and the software that has been developed for architectural design, the consequences of this evolution are obviously important. In a text recently published in conjunction with a NAi workshop,7 Lars Spuybroek (NOX) writes, 'Maya is the most integrative tool available today. Students can combine typical data analysis from programs like Excel (Microsoft Office) with image manipulations from stills or films from Adobe PhotoShop or Premiere and the amazing surface modeling tools in Maya. The drill [the students were given to do in the course of the workshop] emphasized time-based tools like Inverse Kinematics (skeletons with bones and joints, generally used to animate bodies such as running dinosaurs), Particle Dynamics (generally used to simulate snowstorms, fire and smoke, or flocks of birds) and Soft Body Dynamics (used for complex material behavior like fabric in the wind, rubber or jelly-like geometry interacting with other surfaces or force fields such as gravity, turbulence and vortex). One cannot overestimate the effect of this type of software on the minds of architecture students.' Spuybroek is nevertheless one of the few to understand that these effects are 'the effects of machines, not metaphors', as Deleuze and Guattari write. 'An organ-machine is plugged into an energy-source-machine: the one produces a flow that the other interrupts. ... Hence we are all hendymen: each with its little machines. For every organ-machine, an energy-machine: all the time, flows and interruptions. Judge Schreber has sunbeams in his ass. A solar anus. And rest assured that it works: Judge Schreber feels something, produces something, and is capable of explaining the process theoretically. '8 It is in this sense that we can almost consider the famous 'schizo' as the prototype of contemporary architecture.
Yet if we interpret everything in terms of machines and the effects of machines, if everything flows and merges, how are we going to get a grip? Here the diagram plays a fundamental role. Deleuze borrows the concept from Michel Foucault, who employs the word in Surveiller et punir (1975) with respect to panopticism. Foucault observed that the panoptical prison had a function that went beyond that of the building itself and the penitentiary institution, exercising an influence over all of society. Stressing the function of these machines, which produced various behaviours, he discovered this coercive action in workshops, barracks, schools and hospitals, all of which are constructions whose form and function were governed by the principle of the panoptical prison. According to Foucault, the diagram 'Is a functioning, abstracted from any obstacle... or friction [and which] must be detached from any specific use'.9 The diagram is a kind of map that merges with the entire social field or, in any case, with a 'particular human multiplicity'. Deleuze thus describes the diagram as an abstract machine. 'It is defined by its informal functions and matter and in terms of form makes no distinction between content and expression, a discursive formation and a non-discursive formation. It is a machine that is almost blind and mute, even though it makes others see and speak.'10
Little wonder then if diagrams have conquered over the past few years an increasingly important place in architecture and town-planning debates, for these disciplines themselves have taken on more and more the appearance of a process and the (built) results influence processes. This evolution is also fostered by the fact that in Foucault's work, the diagram presents a very clear architectural and machinelike dimension from the outset. The analyses of the Downtown Athletic Club that Rem Koolhaas puts forward in Delirious New York,11 as well as his examination of Arnhem's panoptical prison, appear to be directly inspired by Foucault.
Naturally, most architects make use of diagrams in order to obtain exactly the opposite effect of Foucault's panopticon, or the 'striated' space that Deleuze and Guattari describe in Mille Plateaux12 as 'the space instituted by the State apparatus'. The two philosophers contrast that space with the nomadic, smooth space that should offer greater liberty to those living in or using it. In recent years then, architects have used all available means to render the space of their buildings and towns as 'smooth' as possible, or at least to suggest that. The building is nothing more than a space that develops in a continuous, folding slope and which ideally is seamlessly joined to the ground. If the OMA design for the Villette park is the oldest and still the most visible example of this idea, the Jussieu library, and the open floor in OMA villas that organizes the space between the various individual living rooms, also enter this category because they impose no particular behaviour on the inhabitants. Here the rolling floor is a significant ornament, a form of architectural metaphor, a part for the whole.
Ofcourse, it remains to be determined whether the freedom that architects imagine they are offering with this concept is a genuine freedom. In his preface to the English translation of his study La Prise de la Concorde,13 Denis Hollier points out, in reaction to the theories of Bernard Tschumi and Jacques Derrida, that the liberty in question is a fictive one, because architectural spaces are, by definition, a part of the social system. The earliest texts by Tschumi were partly based on Hollier's work. To avoid any misunderstanding, the English translation is therefore entitled 'Against Architecture'. Nevertheless, Hollier's critique has broader significance and is not only a response to Tschumi.
Datascapes
In the schema developed by Bijlsma and his associates, the MVRDV Datascapes tend more to analysis and representation. They are visualizations of laws, rules, norms and statistical probabilities, and as such they constitute representations of what the sociologist Anthony Giddens calls 'abstract systems',14 that is, bureaucratic systems where the trust in the system as well as the people, institutions and machines that represent it, lies in one's confidence in a certain specialized expertise. In reality, these Datascapes show that the space around us is virtually shot through from the outset with the dominant forces of society. In a single design, there are several abstract systems at work. These systems nevertheless indicate the maximum limits within which the architect can produce his designs. Once the different Datascapes at work have been brought to light, the design becomes the subject of a negotiation in which the architect plays the part of an intermediary and director. Even if the density of legislative and regulatory norms exercises a powerful influence over the design, it is not true that it springs automatically from the accumulation and interference of different Datascapes. To borrow Deleuze and Guattari's expression, Datascapes are not abstract machines, but rather 'bodies without organs' onto which, in principle, each idea can be projected. Similarly, there are instances in which the authorities and assimilated powers impose an invisible discipline on a space. In the final design, MVRDV exhibits the greatest possible number of Datascapes, which are not necessarily in agreement; when all is said and done, the approach is more of a refined form of deconstruction than a unifying technique, in which the design apparently takes shape in the margins. 'Apparently', that is, because MVRDV conserves a secret diagram somewhere that really generates the designs. And if the margins are considered a locus of freedom, Hollier's critique is applicable once again.
In the installation 'Metacity/Datatown', Datascapes seem to function as machines that generate architectural designs for cities. Metacity/Datatown shows how transformations of collective behaviours lead to a transformation of the constructed landscape. The growth in population density of this imaginary city, moreover, remains an important instrument because it reduces the margins. But clearly it is also a form of rhetoric since it refers to a disturbing future scenario in which the population increases so much that it is necessary to take rigorous steps. And the inhabitants of the Netherlands, a country that is heavily populated, fear this scenario of the future more than any other place on the globe.
Consequently, Metacity/Datatown is above all a didactic tool that forces us, viewers and potential inhabitants of the city, to make political choices. These choices may eventually lead to models that have nothing in common with the models presented in the installation. In this regard, like MVRDV's earlier designs, Metacity/Datatown is first and foremost a reflexive design, as Ulrich Beck and Anthony Giddens understand the term, namely, a democratic design that explicitly grants the socius an active role by confronting him with social risks in the form of future scenarios.
The project '3-D City', which Winy Maas recently realized with the Institut Berlage, explores the limits within which a highly dense and compact situation is possible in a series of large architectural models that examine specific commissions for the city. In this commission, however, MVRDV's secret programme is becoming increasingly clear. Suddenly, the piece is no longer a mere didactic design that forces viewers to choose. Indeed, Winy Maas openly describes it as an attempt to create a utopian city in order to anticipate the problems posed by both the increase in the world's population and the protection of the environment. The EXPO-pavilion in Hannover also revealed this change since the building was designed in such a way that it passed for an isolated fragment of a large city that remained to be constructed. In this regard, it was a prototype, recalling the experiments carried out by the Japanese metabolists in the 1960s.
From the start of the presentation of Datascapes, Winy Maas consciously established a comparison with some of the famous spectacular designs in the history of architecture. At that point, the work once again amounted to a rhetorical process aiming to show that it was not absolutely necessary to draw first on imagination and individual creativity in order to realize spectacular designs. In no way was daily reality an obstacle to creativity; on the contrary, daily reality could prove sufficiently spectacular if one opened one's eyes to it.
In '3-D City', however, the intention from the start is to create a spectacular city mixing images of Fritz Lang's Metropolis, Archizoom's Superstudio, Archigram's Fifth Element and the cities of science fiction and Hilberseimer, without Maas's seeming to worry about the intentions concealed by these images or the public's reaction. A certain ambiguity thus continues to weigh on the question of whether the '3-D City' design constitutes a (makeshift) pragmatic solution for a spectacular problem, a radical extrapolation from an existing situation altogether in keeping with Superstudio's 'Twelve Ideal Cities', which contains the essential critical message of a utopia to be realized with pragmatic means. A large number of ideas and values associated with this utopia remained unspoken, not to mention the paranoia of what would happen if we refuse to explore 3-D City. MVRDV's secret diagram appears increasingly like an indeterminate utopian city having an unprecedented density. But what kind of behaviour is this machine going to produce? The exploration of the physics of the constructed environment, which MVRDV inaugurated with Datascapes, seems to turn into an exploration of what Alfred Jarry called 'pataphysics', that is, the disturbing, surrealist physics of the possible. And this is precisely the physics of the schizophrenic that serves as a basis for Michel Carrouges's bachelor machines and the desiring machines of Deleuze and Guattari. As I see it, it would be preferable to undertake a critical evaluation of the 'physics' that Datascapes temporarily provide. We might perhaps deduce the values that would be useful in the next stage. Admittedly, the debate has only just started because most architects and critics unduly persist in viewing the 'Datascapes' design only as one of the many mini-theories that today's architects put by in order to justify their work. In that context, the '3-D City' design may be a necessary, though risky, provocation, since exploration of Datascapes is in danger of getting lost in an increasingly strong demand for quantities and intensities.
The architect's dream
Ben van Berkel, Caroline Bos and UN-Studio, on the other hand, seem to take the opposite approach. In their editorial published in the special issue the review Any devoted to the diagram,15 they explicitly describe the diagram as a 'loophole in global information space that allows for endlessly expansive, unpredictable, and liberating pathways for architecture': 'The end of the grand narrative does not mean that architects no longer dream their own dreams, different from anyone else's'. As they affirm in Mille Plateaux, Deleuze and Guattari demanded of their books, concepts and diagrams that they help to maintain a rhizomatic relationship with (parts of) reality. In other words, they had to use their roots in order to forcibly draw from the world their nutrients. With Van Berkel & Bos, this relationship is radically reversed. Like the architect's dream, the diagram is projected onto the world. Readers will note that implicitly the ideal of freedom expressed here does not concern inhabitants/users but rather the architect. Obviously, this is a polemical stand taken against the architecture of Rem Koolhaas, MVRVD, Christiaanse, Neutelings, West 8 and many others, which Van Berkel & Bos consider as 'pragmatic' architecture. In reality, the situation is more complex of course. Certain UN-Studio's designs, for example, are based on extremely detailed statistical and quantitative analyses; in other instances, ideas are integrated in an extraordinarily refined, precise manner. Nevertheless, unlike the 'pragmatics', UN-Studio favours a formal, aesthetic and metaphorical treatment of analyses: literally even, as when Van Berkel & Bos put a portrait of the 'Manimal' on the cover of their book Move. This computer-generated image, in which human faces and animal heads merge as in a dream, replaces the symbolic figure of a man drawn in a square and a circle, once imagined by Leonardo da Vinci. Whereas Da Vinci's diagram is the symbol of humanism and the central place that man, in this view, occupies at the heart of the world and the cosmos, this 'Manimal' is the symbol of a 'posthumanism' in which all possibilities merge with one another. Yet in Van Berkel & Bos's designs, as in the work of numerous contemporary American architects, this posthumanism does not in any way change the position and role of the architect. Just as the architects of the Renaissance, in the eyes of art historians, seem to have first of all expressed symbolically a view of the world through their constructions, Van Berkel & Bos appear to retain such a privileged role for the architect. However, if everything merges, if there is no longer either a beginning or end to discover, the architect's role is necessarily effected: how is one to lend henceforth symbolic form to reality here on earth and, by doing so, resolve and reconcile breaks, oppositions and conflicts? To fill that role, the architect must turn to a higher order, which remains hidden to most mortals but which he can make visible, be it with his arse, like President Schreber attracting heavenly rays in Deleuze and Guattari's work.16
To that end, Van Berkel & Bos -along with many others- use diagrams that are notably borrowed from genetic technology, chaos theory, complexity theory, string theory, etc. How this theoretical physics is integrated in the design is hardly different from the way in which metaphysics was symbolically translated in past architectural designs, if we are to believe the classic interpretation proposed by historians of art and architecture. The desire to express in architecture the fundamental outlook of a period is naturally a respectable conception of the discipline. Apart from that, popes and priests are not enough for metaphysics, it also demands architects and artists. Moreover, chaos theory, complexity theory and string theory are extraordinarily interesting, and are studied in depth and scientifically grounded. Nevertheless, these theories have yet to teach us anything about the way we behave (or ought to behave) as individuals in everyday life. At most, chaos theory can explain the behaviour of a large-scale population by analogy with equivalent populations. On the other hand, we still do not know at what size those populations ought to maintain themselves. On the purely scientific level, these theories offer no manuals for specific situations. It is improbable that they can, for example, suggest what behaviour one should adopt when an architectural design faces irreconcilable interests. The other way around, these theories offer critics no indication of how they should judge designs. If it is true, for example, as Sanford Kwinter pointed out during a lecture at the Institut Berlage,17 that the entire world, indeed the entire cosmos, is vibrating and that these vibrations determine everything, according to string theory, then by definition any construction obeys this determination, whether it is a building by Rem Koolhaas (as Kwinter, in this instance, would like rather surprisingly), Ben van Berkel, Daniel Liebeskind, Rob Krier or some talented unknown.
Of course we are fooling ourselves if we imagine that a certain type of architecture or town-planning whose forms are in motion can offer greater freedom, and thus contribute to preventing conflicts, the idea being that these forms adapt more naturally to certain flows. If this is the way we must interpret these designs, as certain architects inspired by Kevin Kelly, for example, believe we must, one ends up constructing a theory that in reality recalls the liberal individualism of a F.A. Hayek, based upon a belief in a 'spontaneous order', an 'invisible hand', more than the Deleuzian critique of capitalism. As the political scientist Colin Bird has remarked, 'I can believe that the state really is an organism and yet deny that it has any moral significance whatsoever. The mere fact that a jellyfish is "organic" does not elevate it to a moral status equivalent [to] or beyond that attributed to human beings.'18
It is no accident if the work of UN-Studio proves a success when it comes to resolving complex questions of infrastructure, organizing, for example, Castells' 'space of flows' in the case of the Arnhem railroad station. Here the different forms of infrastructure and the movements are statistical data in a way. Indeed, it is not difficult to translate the flow of passengers into a flow of data. Theoretically, freedom is at stake here, on the individual level, since different possibilities are available for changing trains, but these possibilities are not realized at the site and, moreover, most can be ignored by the mass of people, in terms of the design. The Arnhem station is a machine that looks like the motor of a Ferrari equipped with shiny smooth manifolds. The way the diagram of the Klein Bottle forms in this design a formal lead that runs throughout it once again presents a meaning that is especially symbolic and magic, so transcending the original situation that one can almost speak of an inversion. The concept of transportation flows suddenly seems like the specific expression of a superior order which exercises its power over all the other parts of the design.
Involuntary (de)construction
Greg Lynn has explained on many occasions how the design for the Korean Presbyterian Church in New York was generated by means of diagrams. How in the computer different 'meta blobs' interacted according to their zones of assigned gravitational force. How they grew and melted together into new forms until they achieved a state of equilibrium. How these meta blobs stood for different programmes, single rooms that merged into one big room with a single surface that incorporated the entire programme. How the clients loved this, because they could actually manipulate the forms themselves, making things bigger and smaller without destroying the coherence of the overall concept. Then he introduced a different strategy. A series of tubes was put on the roof of an existing building that we hadn't seen before, the ancient Knickerbocker Laundry. The tubes grew and developed into a rib-like structure with an inner and an outer skin. Tubes were added for access and circulation. In this phase, the smoothness of the blobs was already partly replaced by a certain degree of segmentation, but everything still seemed to be melded together. After that, there must have been a third phase in which the project was adapted to the building methods of the contractor. Constructions appeared and an industrial facade was introduced. In this phase, the project lost its initial smoothness. Today, it looks almost like a deconstruction of a blob.
What at first seemed to withdraw from language, became language again; the diagrams were reappropriated by language. Today, all the materials suddenly tell a multitude of stories, about what they are, how they are made, how they are put together and how they relate to other materials. What appeared initially as a coherent form informed by all kinds of complex systems, suddenly became a complexity again. And it goes further than that because this is a church-factory, a religion plant, which hosts services for 2500 people. It simultaneously accommodates multiple non-sectarian programmes in 80 classrooms, a 600-seat wedding chapel, various assembly spaces, a choir rehearsal space, a cafeteria and a daycare centre. Imagine all these people here, individuals coming from different places around New York, carrying memories from Korea, moving around, doing things in different constellations, like a giant ant farm.
Now, in itself this is not a problem, because as the building stands there it is in some ways maybe even more convincing than if it had been a smooth blob: that would have made a much more disturbing, science fiction-like effect. It would have appeared as if aliens or at least something from 'out there' had just landed. Of course, Greg Lynn himself likes these references and he has referred to B-movie blobs on many occasions with a certain perverse pleasure. Because however much disgust and queasiness they may inspire in movie audiences, they also seem to possess a higher form of intelligence. 'The term "blob" connotes a thing which is neither singular nor multiple but an intelligence that behaves as if it were singular and networked, but in its form can become virtually infinitely multiplied and distributed.'19 This is an interesting metaphor for a building, because a building is never just one thing and is always caught up in a constantly changing, complex web of relationships and stories. That is what makes architecture so fascinating. This process of change doesn't even stop when the building is realized, but continues forever. After it is realized, it is appropriated by the people who use it, for example. I remember a lecture by Peter Eisenman from a long time ago in which he talked about one of his early houses. When it was realized and the clients came for their first visit, the wife exclaimed, 'But I thought we were going to get a Heidi-house!' They first moved into the basement and from day one changed the house and slowly inhabited it until they felt it was their own. Eisenman appreciated that, as he had consciously built in the house a certain resistance. In a way, how the engineers and builders dealt with the original schemes for the Korean church and adapted them to construction methods that they felt familiar with is probably not so different from Eisenman's anecdote -except that it happened even before the realization.
Lynn, however, has always criticized such an approach, or at least he has criticized a deconstructivist architecture that lives on such conflicts and exploits them in geometrical conflicts. Instead of that, he proposes an architecture that is malleable, fluid and supple, to accommodate and integrate all these conflicting forces in a new whole. 'Complexity involves the fusion of multiple and different systems into an assemblage that behaves as a singularity while remaining irreducible to any single simple organization.'20
The building becomes part of a larger ecology and changes with that, which is made possible in the design phase by the latest animation software. In the end, a form is chosen that is static - however, static like a sailing boat, which has a form that allows it to perform well in many different situations. It incorporates all these situations and the final form is mediated between them. In the case of a boat, one could make it more comfortable or faster by changing the parameters, and in the same way one could change the building according to the client's wishes.
But of course, the basic question is what different parameters are selected to play a role in the original ecology? How complex is this ecology really? Who makes the selection of the forces and on what grounds? And, last but not least, couldn't it be that geometrical conflicts in the form of fractures and ruptures are essential to certain ecologies?
In the case of the Korean Presbyterian Church, the ecology still seems to be quite simple. That is not so strange, as it was one of Lynn's first experiments with this way of working. In the first phase of the design process, a software was chosen that allowed one to locate different parts of the programme, let's say the different chapels, the altar and the choir, into 'meta blobs', which grew together. Now their sizes and relationships could be altered, while they remain related and the overall design coherent. Then the original building was introduced and the model roughly adapted to that‹roughly, because they still appear as separate entities. In later, similar attempts to introduce a new organization into an existing building, like NOX's design for the V2 Lab in Rotterdam from 1998, the relationship between the new and the existing form seems already more fluent and integrated. However, it is exactly this limited initial ecology that makes Lynn's finally realized building appear as a deconstruction of the original diagrams. These are almost completely hidden in the final construction and detailing. In that sense, in the completed Korean Presbyterian Church, Lynn comes close to the loosely layered way Ben van Berkel and Caroline Bos use diagrams in their designs as 'interactive instruments': as a kind of mission statement in the management of whole projects rather than something that should be realized literally. In an old piece on Van Berkel's work, I spoke of them with a reference to a text on the oeuvre of the Italian painter Francesco Clemente -which Van Berkel always admired- as a kind of 'diagrams in costumes'.21 The dream of the architect is buried in the whole -that is at one and the same time much more interesting than just that dream.
However, Lynn seems to be more ambitious than that. Much more than Van Berkel, whose work is produced in an innovative yet traditional practice that deals with real commissions, his work springs far more from a tradition of academic and theoretical research and should be evaluated as such. In all his projects, Lynn chooses his own parameters to work with. In the House Prototype in Long Island, they are the topography, the wind and the noise from the nearby road, for example, and in the H2 House for Vienna, it is the light from the sun and the cars from the nearby highway. He also selects a particular software and recently, in his Embryologic Housing project, he opted for a production method as a starting point. As almost scientific experiments in a controlled environment, these projects are incredibly valuable and already influential among a broad group of architects. However, the question is whether that is the sole reason Lynn allows for only a selected number of parameters in his projects. It could also be that his desire to produce a coherence in the design in the first place gets in the way of realizing this coherence in the final building, because the real ecology in which that finds itself is much more complex than Lynn's selection of forces that play a role in it. 'Any object supposes the continuity of a flow, any flow, the fragmentation of the object', write Deleuze and Guattari.22 Fearing the fragmentation of the object in the reality of everyday life, architecture withdraws into a body without organs. Michael Speaks could be right when he says that Lynn, like his mentor Peter Eisenman, is still too much interested in the metaphysics of architecture.23
Machine effects
Unlike MVRDV, whose work makes use of the diagram above all as a integrating form of notation, a formal abstraction of a given complex reality, and Van Berkel & Bos and Greg Lynn, for whom the diagram is a chosen form that generates and structures the design, Lars Spuybroek is alone in conceiving the diagram as a complete machine. 'The diagram is a very clearly lined network of relationships, but it is completely vague in its formal expression', writes Spuybroek in 'Machining Architecture'. 'Diagrams love pulp, and they only recognize materials at their most heated and their weakest stages. The diagram is basically a conceptual input/output device which swallows matter and, while restructuring, also ejects matter. In that sense, every informational plane is always an interface between material states.... The diagram... is an engine, a motor: it doesn't want to impose itself on matter, but to engage in a process of continuous formation‹it operates at the backside of the image, on its blind side. Diagrams are the informational nodes and codes of the world; they are stabilizing contractions in material flows -first they channel and then they relax. They are faces in a landscape, singular perceptions connecting streams of actions. They are lenses, mirroring a movement: first a contraction of matter-energy onto an organizing surface, then an expansion into many new other structures.'24
Contraction is the phase during which information is collected, selected, converted and graphically organized in a virtual machine. It is a process in which a three-dimensional network is converted into a two-dimensional surface. Spuybroek describes it as a movement towards quality, order and organization. Then there occurs a process of expansion in which the machine, the diagram, is put into the material, spreads throughout and gives it a shape. This is a development that enables one to shift from a two-dimensional surface to a three-dimensional structure. Spuybroek describes this as a movement towards quality, materials and structure.
Up to this point, there is nothing in this approach specifically involving a computer. In principle, an expressionist method of working, described by Vassily Kandinsky, for instance, as the chain Emotion-Gefühl-Werk-Gefühl-Emotion, is equally suitable.25 A process of contraction (Einfühlung) and expansion (Expression), which is endlessly repeated, is also at work here. In that case, individual people are themselves the machine. This process, however, can only be active in that it is a process that is transmittable to others, as if there existed an agreement about the nature of human beings, and all men corresponded to a humanist ideal. This is problematic if we accept that individuals are part of flow networks, in which they must endlessly make choices, and that these choices continuously modify the position, identity and therefore nature of these individuals. The process then becomes altogether subjective. In the architecture of UN-Studio and Greg Lynn, that subjectivity is not resolved because they choose diagrams that exist outside the process to generate their designs. Spuybroek's intention then is to free as much as possible the design process of the architect as well as the individual person by constructing an abstract machine that also lies outside himself and which is linked to the world rhizomatically.
As Spuybroek sees it, the computer is only a machine that reinforces communication between different diagrams. It is a diagram in and of itself, and the specific computer programmes are also diagrams in and of themselves. For Spuybroek, there is no difficulty in describing as 'material computers' the dynamic models borrowed from Antonio Gaudí and Frei Otto, which he uses in certain designs to determine his buildings' form of construction. These are material diagrams because a change in one area influences the form of the overall project. The design process comprises then, in the end, a chain of different diagrams forming a design machine since they are coupled to one another and are continuously converted into one another. For this reason, Spuybroek believes that schools that teach computer-aided design should offer as well instruction in 'computer-aided conceptualization' and 'computer-aided manufacturing'.26 Even if the computer is an expression of a tendency to make everything smoother, more fluid, Spuybroek shows that there still exist couplings and connections that harbor within them possibilities of choice. The 'space of flows' is not a fatality: We can still manipulate it in our own way.
Since then, NOX has developed this way of working in a series of designs and completed projects that include 'Off-the-Road-5 Speed', a housing design for Eindhoven in which the five speeds refer to as many diagrams; 'Wetgrid', an idea for the exhibition 'Vision Machine' held at the Musée de Nantes, or the D-Tower of Doetinchem.27 The D-Tower design assumed a form that is, in the end, both real and virtual, for, apart from being a real object, the tower itself is also a new diagram which continuously transforms the population of Doetinchem's input to an Internet site.
What is interesting yet problematic in NOX's way of working is that the result (the design and buildings that arise from that) is almost impossible to represent. It is interesting because these designs normally permit a large amount of interactivity with the future inhabitants and users of the building, or because they may play on the modification of subsidiary constraints until a relatively advanced stage; it is also problematic because the design proves difficult to present to a client, for example, who requests that the final form is visualized before the last stage of the process. Thus, up to this point Spuybroek has relied on clients who have been able to reflect at the high level of abstraction of the design process and interpret diagrams on their own. If we wish to consider the computer as a true tool in design rather than a mere implement, if we want to know to what extent we are manipulated in the 'space of flows', it is unavoidable that we resign ourselves to this method. But then again, if Schreber was able to theorize about himself, there is no reason why we cannot as well.