back to WRITINGS
I Am Trying to Imagine a Radical Free Market Urbanism
Conversation between Peter Eisenman and Patrik Schumacher, New York 2013
Published in: Log 28, Stocktaking, Summer 2013, Anyone Corporation
PETER EISENMAN: What is interesting in Mario Carpo’s book is I’m at the beginning and you’re at the end. I don’t feel that you were where I was headed. My work has always been about the possibility of the unauthored, the possibility of complex adaptive systems operating in a certain way. So, yes, at a certain point in time in the mid-’80s I was interested in these ideas. A lot of what Greg Lynn has come to do came out of the work we did in our office. Greg is doing an exhibition on the archaeology of the digital where Frank Gehry and I are seen as the precursors of computational logic. We were drawing by hand things that computers do. Jump to Patrik Schumacher, and you make the most complete statement, I think, of where the parametric, or computation, lies in your two books. Didactic, comprehensive, persuasive. We don’t want to rehearse and go back over those two books; what I want to do is ask some questions which are raised for me in the movement from the beginning of Carpo’s book to the end. One is the question of how you choose. In other words, if there are primitives; if they are developed by algorithms which have no knowledge of architecture or the discipline of architecture, what constitutes scale, what constitutes place, what constitutes materiality, i.e. things which are architecture? There has to be moments in any system of authorial choice. There is a style, yet it’s not the style of the computer. It’s a style that is different from Hernan Diaz Alonso, different from Greg Lynn, different from people who would also argue they’re using computation as the basis of the work. From the universe of possibilities and x power of options that are presented by computation, how do you choose?
PATRIK SCHUMACHER: Well, certainly, it’s true that you have to choose. First of all, you have to choose a setup, and a set of primitives, i.e. a slice of the universe of possibilities. And often when we start, we have several slices we’re investigating, but it’s true that then you select and perhaps reset and rerun – it’s iterative of course. I think there are two sets of criteria: there’s the functional criteria, which allow you to fit the program, fit the site, and so on; the other is the formal criteria, the aesthetic criteria. And they are initially quite intuitive. You can’t rationalize them on the spot but I think you can rationalize your intuitions when you step back and critically reflect on the principles that govern the intuitions you follow. But in the moment of design work you make an intuitive gut choice, i.e. you apply your aesthetic sensibility. Sometimes with different projects you look for different nuances, something that is more tense, more complex, more aggressive, or something more calm. We make these intuitive choices.
PE: But you also have to reproduce your style, otherwise you would not recognize a Schumacher/Hadid project from a Ben van Berkel. But if you are making intuitive choices from a limited range of possibilities, why is that different from architects who don’t work in a parametric process, who don’t work with primitives, who don’t work in the way you do?
PS: The aspect of stylistic identity is being preserved. What is different, is the range of options you can choose from. You can produce and reproduce much faster, and much more. I also believe that what you can choose from has more intricacy and complexity and more order than what you would be able to do by hand. Some of the more complex, differentiated fields, or the correlation of multiple subsystems, each differentiated according to its own independent logic, and then made interdependent via associative logics, cannot be handled by hand. You simply couldn’t do that before computation, you couldn’t give yourself the “rules” and adhere to them rigorously. Its just too complex. Or, if you tried to do it, it would literally take weeks to generate something which in the end still would not have the same coherence as a scripted arrangement.
PE: Let’s take your project for Istanbul as a special case. It seems to me if I didn’t see the surrounding area I would say to myself, hey, this is a nice project for 20 hectares or so. But when it’s 400 hectares, or 600 hectares, I say my God. This is no longer a complex fabric when it’s at this scale, it’s a unicum. My question is, how do you make this part of the fabric different from this part of the fabric, because it looks like the same thing gets repeated. When it gets to be a certain scale, it doesn’t seem to be what I would call a complex adaptive system; it becomes a holistic system which doesn’t allow for smaller-scale complexity.
PS: You’re just looking at one layer, you’re just looking at abstract lines. You need to look at how it’s developed into a fabric. This is still an abstraction. You also need to compare it to what a modernist would have been able to do. First of all, in the studies we show how we can establish morphological continuities from the existing urban fabric into our fabric, and then there’s a rapid modulation and transformation of these adaptive modules that produce a very large range in scale, from a solid block to a courtyard block to a split courtyard block to blocks where the corners pull up into towers. And then there’s a second primitive, which is the cross tower. So we’re starting with two ontologies of fabric. Each of them has a series of modulations and transformations, and in the end they transform to become formally close to each other. We have a cross tower and then a block where the four corners are pulled up to create a pseudo–cross tower. In that way the two ontologies become affiliated to each other. Each has a large range of manipulation, which bleeds back into the existing fabric. That’s the versatility and complex adaptive capacity of our Istanbul urban model. Thus there is a sense of adaptation to surroundings, internal coherence and modulation so that you can follow gradients. There’s a lawful transformation. You have maybe five or six or seven variants but they’re all in a sequence that you can follow as a trajectory for navigation, what I call the vector of transformation, and it leads you into an existing condition with which it makes affiliations. It continues so that our figures actually bring in the existing context. There’s another ambition here, namely to establish local to global correlations that allow for local to global inferences that can further orient those who are trying to navigate the field. Where the global shape of the field is widening the fabric height is increasing. Thus the global plan is mirrored in elevation, which becomes lower in the narrow point and high in the wider zone.
We also defined similar adaptive logics in our Singapore master-plan. Here we were faced with a large site that was surrounded by disparate contexts. Instead of adding one more incongruous texture into this collage we tried to establish formal continuities in all directions which in turn generated an internal complexity. We created a new complex unity that was much larger than what was defined by the boundary line of our allocated site. Our morphology radiated outwards because we had allowed the outer existing context to radiate inwards. Compare this to modernism, which is blind and insensitive to the context, a kind of monotonous imposition. Or compare this to what happens in most of our contemporary urban developments, or in the postmodernist collage, where you just parcel out these sites and let everybody do whatever they like. The result of this would be an absence of identity, because in the collision of pure difference, what you generate is something that is akin to a garbage spill. In each garbage can there are very different things, but once they are all spilled out next to each other, they become indistinguishable white noise. You may always demand more differentiation from me, and yes, I would love to do more, but you also have to see this method and this result in comparison to what has been achieved elsewhere.
PE: So you’re saying at any scale it works. There’s no scale that is too large for you.
PS: Yes. I would not think of it as a single authored work, but we could go very far with it because we have this internal repertoire and richness, we catalog hundreds of ways of modulating and hybridizing and so on, so I wouldn’t be afraid to take on something much larger. We had a certain reluctance earlier, for instance, when we were given a 2 million square meter, five hectare site or something like that. We were doing a master plan and we thought we should invite other architects, but the client said, “No, why? Just do it all.” So it was initially counterintuitive. But then we worked on it and we generated several typologies – we did this also in Beijing – and each of them was varied and recognized different functions, and for the different functions, several typologies, and then we ordered them. I thought at the end, maybe you can do it. Ultimately for larger sites, you can bring in other authors and architects, but I would say they would have to be working within the methodology of parametricism, i.e. there would have the shared ethos of creating affiliations and continuities as different ontologies are being build up. Then I imagine you could create something powerful and viable, an open system with multiple authors, who share a unified style with a shared methodology, the style and design research paradigm which I call parametricism.
PE: But there are a lot of people working with computation that don’t produce things stylistically the same.
PS: You could say that our style is different from van Berkel, from Greg Lynn, from Jesse [Reiser], and so on, but I think it’s not.
PE: No? I can recognize the differences in each.
PS: You can recognize, but oftentimes that is more due to the inherent limitations of a firms preferred techniques, and not a sought after, desired result. I can certainly recognize and subscribe to and own up to Jesse’s work, to all of it. To some extent I can criticize his work the way I criticize our own work, and I think we’re sharing the same principles – this is what I’m trying to formulate in the heuristics of parametricism – we share a set of values and criteria: the demand for more differentiation, and at the same time for coherence, adaptive capacity, and continuities. We all we share this. And I think we also now share not only the desire to be multi-system but for these multiple systems not to just be colliding into each other but to intricately interweave and inter-articulate, so we all share similar methods and virtually the same aesthetic sensibility. And if the works of these authors nevertheless assume different, recognizable identities, it’s not due to a different ideology. I think it’s the limitations imposed by the particular tools used and the particular history of projects. These are just accidental limitations, which we shouldn’t overestimate. If clients want a particular, personal style or architect’s handwriting, that’s a different question altogether. I’m not interested in these individual styles. What is important is that we all share a radically new, original paradigm that makes our work interesting and relevant to each other. And one more thing: The phenomenological similarity of all the results of parametricism exists only in its shared contrast with all prior, static, discrete architectures. Once we move beyond the general, abstract characteristic of organic fluidity we recognize the incredible, unprecedented internal richness of this style’s output: a huge variety in comparison to which all prior productions put together seem repetitive and narrow.
PE: Several questions come out of this. You could argue that most great cities, 80 percent of them are unplanned, that is, they don’t have the control that you have in these large areas, and it seems to me that when you work at this scale there’s no relief from it. In other words, there’s no dumbness in the system. It is the hypersaturated with an attention necessary to the work that I question at that scale. Should it be totalizing? Not just different architects, but shouldn’t there just be voids in the system where something else takes over, a developer who just puts up junk?
PS: No, I don’t think so. That would destroy legibility, continuities. Can you bring in spatial singularities? Yes, you can if the programme merits this. However, such singularities need not be disruptions. Rather they should be culminations or points of conspicuous intensification within a field. Points that radiate through the field and announce themselves this way. I am aiming for beautifully intricate urban textures, that seem like natural systems. All these design processes we are using are originating in simulation tools for natural processes of self-organization. Let’s say you have a rock formation that has eroded over centuries, then you have moss growing on it, picking out and selecting certain areas, and then you have larger plants growing out of this, and these are more on the south side while the north side is still naked stone. For me this is a lawful buildup of correlated subsystems, and that’s what we are attracted to and try to emulate in our computational models. We intuitively sense the order and performativity of such correlated multi-system set ups – that intricate order is what we recognize as beauty. Although each era has its own specific aesthetic values, there is a universal aspect in all aesthetic valuations: the search for order and the rejection of chaos. The difference is that we can now recognize and produce a more complex forms of order, an order that would have appeared to be chaotic to older sensibilities. You can come in and make a random interruption into such intricate orders, you can maybe put in a disruption or gap that draws attention to itself in an urban fabric. But how many times can you make an interruption before order disintegrates into visual chaos? If you want to add to or intervene within an intricate order, my methodology demands that you try to formulate and follow a rule of intervention, to weave a new, enhanced, more complex texture. That means that isolated interventions are excluded. All design moves must be rule-based., although there is no limitation with respect to the creation of new rules. Moreover, my methodology can weave integral textures out of disparate inputs. Give me any collage of initially unrelated elements and I can generate connections, resonances, invent correlations. I reject the pure interruption, the pure discontinuity, collage. That doesn’t mean I’m not craving for as much versatility and diversity within this coherent texture.
PE: What about charges of vitalism, of animism, the charges that people bring against these adaptive systems that sort of mirror nature. That doesn’t bother you?
PS: No, this is a misunderstanding. It’s not for the sake of mirroring nature, not for generating the picturesque.
PE: No, but it’s a critique that you hear.
PS: But I can argue for these methodologies and for these values purely on sociological, instrumental grounds. That’s what I’m doing. I don’t validate it because it’s like nature as if this would give it a halo of profundity. I have to reject that. We don’t need that. Our ways of working can be justified in relation to what I identify as the societal function of architecture.
PE: Are these cities, are they of this time? Would you say that this is the work of this time? That it is the idea of what the urban can be, given the processes of computation, algorithm, etc.? That we must use these functions in order to produce the city of today?
PS: Yes, because I believe that it’s life enhancing and that it enhances social processes. So it’s not only recognizing the new opportunities – that we can built more complex now – it also addresses a vital societal demand.
PE: What is that societal demand?
PS: The societal demand for a complex, communicating environment. We are living in the post-Fordist network society. We live in a knowledge society, a society where the cycles of innovation are much faster. The intensity of communication has increased massively. We are all working on projects where we need to recalibrate what we are doing with what many others are doing. That’s why we need mobile phones and the Internet. That’s why we need more and more face to face communication, and why my work is 100 percent meeting, meeting, meeting, meeting. And for most of our ranks now it’s the same. Our work is primarily communication, the bottleneck is communication, making communication effective and intensive. Literally less than one percent of the population produces all the food in America, and in Europe it’s the same. Go to a BMW factory, they’re run by robots with a few people just monitoring. Most of the staff of such companies work on research, design development and marketing, and that’s happening through communication, trying to coordinated so many interdependent aspects in ever accelerating cycles of innovation.
The built environment, for me, is an interface of communication. We need the built environment to unfold continuously well-ordered choices so you can navigate, orient, and quickly find other relevant communication partners. The built environment needs to sort, distribute and order a myriad of different communication scenarios. Order implies orientation, the ability to make inferences from the seen to the not yet seen. In contrast, environments that are constituted by the endless repetition of isotropic sameness, in the monotony of modernism, you don’t know where you are or what’s where, you cannot make inferences, because the programmatic differentiation of today’s society cannot find any registration. Orientation and inferences are also impossible in the visual chaos of what I call garbage spill urbanism. The urban spaces we are designing try to create a navigable order and they try to maximize visual penetration: they are immersive spaces where you come in and have layered offerings in all directions, in front, below and above, and layered deep into space. Such a space makes hundreds of destination choices simultaneously visible and indicates in which direction even more choices will become visible.
PE: Let’s say I accept what you’re saying. Why then the formal aspect? Why do they look the way they look? You’re saying it does all of those things – the communicative possibility – and I say to myself, why should the possibilities look like this? Why is their form important?
PS: The problem is, first of all, you have to achieve complexity of organization. So you need to have density, and a multitude of different things brought closely together. So it can hardly look like one big simple form, like a clean cube or cylinder.
PS: Because then you don’t give the specifically required space to the multitude of different functions. The resultant morphology, when you just give space to the requirements in a pragmatic way, would come to look like the collage or garbage spill of the postmodern city. But I say it urban space does not only function through physical distribution, through the spacing out or making proximate. Built environments function only through being navigated by perceptually engaged, active agents who have to find their way. So the curve-linearity we privilege, as well as the gradient differentiations we privilege allow the urban scene to become complex while maintaining legibility. They have to be complex because the stuff to be accommodated is complex, but you have to make it perceptually palpable. You need to make it possible for the complex scene to be perceptually decomposed and navigated. This is what I have been calling the design task of phenomenological articulation. And more recently I’m also working on developing a semiological system with these designed environments. The engagement with phenomenology recognizes that users are cognitive agents. The engagement with semiology recognizes that users are socialized actors who read the environment like a code that allows them to modulate their behaviours in a coordinated way when they cross spatial thesholds. Users are primed to participate in specific communicative situations as they enter the specifically designated and coded spaces. The social functionality of architecture thus operates via its perceptual tractability and via its semiological comprehensibility. That’s why it has to look in a certain way, namely to be spatially sufficiently complex, visually decomposable and communicative. Our aim is to create information-rich environments. This is not what the global movement of parametricism knows itself to do already, but I’m saying parametricism has the capacity to do it. This is my attempt to push the style forward via new explicit ambitions and relevant criteria of success. It can be done, I can do it. I try to do it. The aim is the design of information-dense, information-rich, navigable environments. As recent built example I could show you our mixed use Soho Galaxy complex in Beijing. But we can also look at our master-plan for Istanbul.
PE: Yes, I want to talk about this for a minute. This image, you could argue, is very similar to Le Corbusier’s Maisons a redents in the Ville radieuse, except the shapes are different. But the layout of the spaces – relative to streets, relative to sidewalks, relative to plazas – seems to be the same except stylistically different. What, then, is different?
PS: Each point in the urban field of our master-plan is embedded in a sequence of transformation that modulates building height, block size, grid density and directionality. Each block is also located within a typological morphing series. So urban dwellers and visitors can navigate the field according to all these gradients, like bacteria are navigating a nutrition gradient. Through local changes you’re drawn into a meaningful global position, and you know your relative position within the smoothly differentiated field; it’s like the smooth spaces of Deleuze and Guattari where you’re navigating transformational vectors. If you look at the detail of our design, you find that there is a deep relationality, in the sense that the local tectonic articulations are integrated into a global field logic. For instance the public face of our blocks shows a deep relief, and then you are pulled into semipublic courtyards where the façade articulation gradually flattens, and the façade becomes a smooth, flush surface in the private courtyard areas. The facade relief is modulated as you go deeper into the interior courtyards: inside, shallow; outside, deep relief. Do you see this contrast?
PE: It’s not clear, by the way, it’s very good that you point that out.
PS: Sure, you might have to be pointed to that. But such correlations might also be picked up spontaneously. After all we are programmed to seek out correlations that make our environment predictable to us. Once we learn to expect such telling regularities in our built environments we might seek out, find and read them more readily.
PE: It is a semiotic system.
PS: Yes, but it’s a parametric semiology.
PE: A lot of people would argue that the parametric semiology, or complex adaptive systems, is a neo-positivism. In other words, the general critique of your work is that it is a new positivism, that it doesn’t allow room for accident, for the arbitrary, for chaos, for discontinuities, and that it is a totalizing thing. Is it a new positivism?
PS: It’s not. It’s an open system. There are the heuristics of parametricism – I call them taboos and dogmas – because you need to have a convergence of a collective group of researchers on a set of principles and values, but these are so abstract that they open up a huge universe of possibilities while they shut out all traditional and modern forms. They are open-ended and radically open to unpredictable inventions and turns of argument. I also believe that parametricism, going forward, has unpredictable possibilities to create new ontologies, and subsidiary styles not yet imagined. It’s very open-ended, like the biological evolution: endless forms, totally unpredictable and ever evolving. Some things you know for sure: it’s going to evolve through the mechanisms of mutation, selection, reproduction, it’s going to be adaptive, correlative, with survival criteria applied to it; it’s going to be DNA-based, most probably; but its concrete results remain unpredictable. In another sense, on a theoretical level, the sociology I’m basing myself on, in terms of [Niklas] Luhmann’s work, is also far beyond positivism in that I am aware that this is not about historical necessity – it is an offering or choice which I proselytize for, which I think is the best we have come up with up to now. It’s not a good reaction to say, “Well, you’re driving the arguments forward relentlessly. Isn’t there any relief? Shouldn’t there be something else?” Show me, put something else on the table, and we’ll debate the relative pros and cons. You cannot just say something else should be allowed in. Tell me what this something is and we can judge its relative merits. My philosophy is not positivism, its radical constructivism. Radical openness coupled with historically up to date criteria of evaluation that demand architectural principles to deliver high performance for the most advanced, high productivity arenas of contemporary world society.
PE: Let’s take [Oswald Mathias] Ungers’ Green Archipelago project. That is, you create, instead of continuities, holes, he calls it the dialectical city. Rem [Koolhaas] has also worked on that kind of idea. Would you argue that your city is following from Ungers, or absolutely a radical differentiation from Ungers’ project, which was a radical differentiation from Collage City, from the collision city of Colin Rowe? I would argue that Rem and Ungers occupy the same space. And they were a model that worked in the late 20th century, let’s say, from ’75 to ’90-whatever, when the first digital revolution came about. How would you compare this project, your project, to the green archipelago as urban work?
PS: I look at the whole of the history of architecture. Its prehistory, then its initiation into the Renaissance, the Baroque, Neoclassicism etc. I always look for the moments of progress and advancement, also in relation to societal challenges and opportunities, and I see value and progress in all the figures you mentioned. And as far as I’m concerned, if there is a valuable new possibility or concept, it deserves to be recuperated, absorbed, enhanced, and integrated into a new project. For instance, I think that Rem’s idea, that urbanism generates potential and that architecture consumes potential is rather insightful. Also, his 1987 urban plan for Melun Senart that focusses on determining the voids and leaves the building fabric indetermined is still very interesting for contemporary society. You could compare that to your desire for rupture, for the unpredictable and so on, and the open-endedness of a process, and the resistance to attempts to determine everything to the last point. I agree with that, that’s in our plan too, coherent with the heuristics of parametricism. How is it in here? Remember John Rajchman’s set up for the virtual house competition? It was published in Any Magazine. It was great, profound. It instrumentalized Brian Massumi’s concept of audience appropriation and the Deleuzian discourse for a powerful architectural methodology. I believe that you could read some of this into a lot of this differentiation we’re doing. We produce a morphological richness that is always excessive relative to its intended semiological utilization, never fully exhausted; it’s enriching the environment with potential for aleatoric appropriation, and it’s stimulating creative utilization. How can you generate true flexibility in a neutral Miesian grid? There is nothing for people engage with? They’ll just fall back upon their routines. One has to challenge creativity through automatic processes that deliver true novelty. You have been the trailblazer of this methodology. I think we have to challenge the user’s imagination, through a strangely structured environment, to see possibilities we could never imagine. It’s like reading Turkish coffee grounds, or like Leonardo looking at amorphous watermarks on the wall and seeing figures and landscapes. You can’t give the totality of your work over to that but I think there is always an element of random mutation. And in our design work, as an important moment of the creative process, we browse through a plethora of graphic structures, select and interpret in relation to the novel, unfamiliar stuff the process throws up. We give new meaning in response to this process stimulation. That’s where creativity resides. This kind of creative reading should be allowed to take place within the building itself – maybe not to the same degree in all aspects of a project – but zones for aleatoric appropriation should exist in the final project. This is also similar to Koolhaas’ notion of urban potential. But for me it cannot be based on emptiness. Instead it must be an over-structured environment, where I haven’t given meaning to it and I can’t predict what meanings will be given to it, or, I’ve given meaning to it, but I am ready to accept that others give other uses to it. Because it’s unusual, it’s strange, it’s stimulating, filled with countless abstract offerings, intensely latent, but nothing manifest.
PE: If we, say, go from Camillo Sitte to Le Corbusier to Frank Lloyd Wright to Mies van der Rohe, I would say that Ungers’ project was an urban project, as is Koolhaas’s project, and it would seem to me that your project falls in this historical line, that it is the next urban project in that line. Let’s accept that may be so. Can you see any other projects today that either challenge your project or are aleatory and in the same line? I don’t see another urban project.
PS: This kind of urban project I’m looking at was partly inspired by your Rebstock study, and also by what Jeff [Kipnis] was doing at the AA in the early ’90s, there were definitely precursors there. Frei Otto also had a strong influence on my approach.
PE: I’m saying now. Not Archizoom, not Archigram, not Superstudio.
PS: They’re just in the background. Superstudio was kind of ironic, but there was an interesting ambiguity between utopia and dystopia and irony. A declared distopia became to fascinating and so obsessively pursuit that it tips over into a positive desire or utopia. I like the boldness of this era.
PE: But where is that boldness today?
PS: We have a lot of young architects working similarly. Not on a specific project, but in parametric urbanism, there are quite a few people picking that up. There is a certain boldness here too. It’s not out in reality yet.
PE: I believe that the world is divided today into two camps. One is the phenomenological camp out of Norberg-Schulz’s Genius Loci, Juhani Pallasmaa, Richard Sennett – the Craftsman. The other is the conceptual, the projects of Rem, Jeff, Pier Vittorio Aureli. Where would you put your project?
PS: Well that’s difficult, I wouldn’t fit into these categories.
PE: Okay, that’s fine. I mean, you don’t fit into them.
PS: I would say for me the phenomenological articulation is a major task of the designer and of architecture.
PE: You’re not a [Peter] Zumthor.
PS: No. What I miss in the phenomenological camp, what I really critisize there, is that there is a sense of making that is just a kind of sensuous, experiential experience for the connoisseur. As far as I’m concerned, phenomenology has to be instrumentalized for the articulation of a high performance communication project.
PE: If it is instrumentalized, then you have no problem with it.
PE: Isn’t the instrumentalization a conceptual project then? To instrumentalize the phenomenological means to bring in the conceptual. I would argue that yours is a conceptual project. It may drag in the phemonological, but you are interested in instrumentalizing affect, the affectual experience.
PS: Yes, but the way you talk about it, when you say conceptual project, I think you are fetishizing the idea. Where’s the innovation? Yes, you can say somebody like Rem has made a series of conceptual innovations through his writings and polemic projects.
PE: Rem and Zumthor couldn’t be further apart.
PS: Somebody else may have been working on the dimension of sculptural presence, phenomenological presence, light, materiality, and so on, but I’m stepping back and saying I need all of these components to generate global best practices to deliver the built environments that contemporary social institutions demand. I criticize your work to some extent because I think you’re a great innovator on the level of concept and process – reflecting process and making it productive – but when I look at your work I feel that you could have benefitted from reflecting the phenomenological dimension better. What I love in your work is precisely the aspect of phenomenological multi-valency, that is the potential for multiple perceptual decompositions or readings, the ambiguity that results from interpenetrating spatial orders you picked up from Colin Rowe – phenomenal transparency – and this is very very strong in your work. And when I lecture on phenomenal transparency and gestalt psychology, the building up of ambiguous figures and so on, I use your work as my preferred architectural examples. But when it comes to the built project, it can’t be just operate impactful enough by pure geometry only. It needs to acquire sense of phenomenological presence that comes with attention to materiality and light. I think we sense our environment not only visually, but with the whole body where we feel lightness, heaviness, and that’s the way we orient and navigate space. I feel sometimes, and this may be harsh, that you don’t do that, that the environments you create don’t have the force required to truly stimulate and you don’t give your structures the material power and force that compels our attention and trust in them as forces to be reckoned with, you don’t deliver sufficient presence. It’s not substantial enough to draw you in. Your works are like stage sets; it doesn’t give me the sense of reality that would compel me to pay attention to its ordering suggestions. Plaster and sheet rock cannot compete with concrete, steel, stone. Even the material magic of carbon fibre compels attention. So its not heaviness, its character that comes with material performances and specific affordances; the different characteristic presences and levels of force to draw you and propel you. These are mediated via phenomenology, i.e. via visual, tactile, accustic as well as proprioception and vestibular perception etc. Initially I’m always going by my intuitions and by what I am feeling, asking why I am attracted to this, why I am exhilarated here; and then I am trying to analyze what it is that works and what doesn’t work intuitively. This way I can rationally validate or critique my aesthetic reactions. But there is a caution to be observed: The architect needs to distinguish and assess the difference between one’s professional sensibilities as designers, the way we read and evaluate buildings as expert connoisseurs and versus how the ordinary users of the buildings would experience them. The purpose must be to construct successful, innovative, productive spaces for users who are in the midst of their high performance pursuits: Spatial orders and spaces that communicate and frame communication on a new level of complexity and intensity.
PE: I would argue that is an excessive idea of presence; that is, we can’t have a world of all literature, of all art, of all high whatever. We would explode. We don’t need Peter Eisenman buildings on every street corner. We don’t need Patrik Schumacher and Zaha Hadid buildings everywhere. That’s what I would argue is the excess of presence. You’ve just denominated that we need more presence and I question how much more presence; that’s one of the things that separate us: how much presence we need.
PS: You can’t have presence, presence, presence, presence. You need to select, accentuate this and suppress that.
PE: Can you imagine a campus full of Zaha? Or Rem?
PS: In the end it will all be relative. You can’t have only moments of heightened tension, one after another. But, by the way, I just remembered, back when we discussed this idea earlier – the idea of the aleatoric – it’s based on your idea of the space of becoming, which I’m very interested in, a space that becomes something else each time an audience engages with it. You talk about the architecture school at Cincinnati, where you start with the grids. If there’s two or three interpenetrating grids one could trace them and could oscillate between them or hold them in simultaneous presence. But what if the simultaneous ordering system proliferate to become five, six or seven. It becomes an interesting, intense field where you have to actively select, pull out, you have an overwhelming ambiguity but you can actively select connections. The decomposition of the scene becomes very unstable, multivalent, and therefore each new audience with its own agenda has a chance to draw out figures, territories or units of interaction you might yourself not have foreseen.
PE: Have you been to the Cincinnati building?
PS: Yes, I’ve been there. It works. I think it works intellectually, as an idea, very powerfully. You need to understand that your idea of a “space of becoming” is a key strand of the paradigm and project of parametricism I’m talking about and that I pursue in our design practice.
CYNTHIA DAVIDSON: Can I ask a question before we close? Peter usually says, when you go to see your shrink, he’ll always ask, “What’s the hot issue for you? What is it you don’t want to talk about, but is a burning issue?” In that context, I would ask you, what is the hot issue in 2013 for architecture, as you are involved in it?
PS: I have two points. One is whether there could be a radically free-market urbanism that has, at the same time, a new level and density of order. I am trying to imagine a radical free-market urbanism, which does not produce the current garbage spill result of deregulated urban development. I am trying to theorize about an architectural discourse that does not rely on or create its order through state planning and heavy central, prescriptive regulation, but should be possible through open-ended interventions by many. The presumption is that unhampered market exchanges afford a more efficient, robust and dynamic information processing mechanism to allocate land and positional resources to their highest valued uses. Planning is faltering on an insurmountable complexity barrier in the face of dynamic, distributed knowledges. All global planning degenerates quickly into a fetter of progress. What can step in here as a cohering and ordering force? My answer to this question: What it requires is a coherent architectural culture, a hegemonic epochal style that is as versatile and adaptive as it is cohering and integrating. Within that, as long as you adhere to the heuristics of parametricism, you’re totally free to intervene and weave unforeseen figures into an ever-evolving, intensifying fabric. This works as long as each new, unforeseen intervention forges affiliations and continuities, selects its articulations according to an overarching semiotic code, and according to principles of phenomenological articulation. This does not require a single ordering hand, but it requires the cohering, shared methodology that is disseminated through convergence of the discipline into a hegemonic style. Under this premise we can I leave the programmatic briefs and land usage totally to the entrepreneurs who’s programmatic investments are allocated and coordinated in the free land and real estate market. In theory, there is no need for political planning, just as there is no need for political planning in the evolution of our internet and telecommunication infrastructure and content distribution. In practice it’s a big challenge. We work on some master-plans and are confronted with the dialectic of order and freedom. We initially felt it is obvious that we have to rely on the planners. We were trying to convince the planners to allow us to regulate the city impose our order and they resist us by saying we don’t know what will be required by the market. The client bodies – like landowner associations – are rightly afraid that any imposition would compromise their ability to respond to market forces. But this worry has also entered the mind of the planners themselves. My conclusion is: I’d love to promote a radically free-market urbanism, but that would only work if all the architects who are hired are bringing their versatile engagement with market contingencies under the spell of parametricism’s methodology. This could work because parametricism’s methodology is a methodology of dynamic adaptation. Rigorous adherence to parametricism would be comparable to the universal modernist way of working of the 1950s and ’60s. Then the discipline had a coherent canon and methodology and set of values where you didn’t have to regulate explicitly, you could just rely on anybody coming in and adding to the modernist city, and that’s what I think would be the solution now, except that under the new conditions of increased complexity and dynamism it can no longer be modernism or its contemporary offshoot minimalism. The second point is the idea of creating a built environment as a system of signification with a much higher level of grammaticality and order – it’s the semiological project I am promoting recently. At the moment I can only think it through as a single author of, again, as large as possible a piece of architecture, like e.g. a university campus, or something like Google campus. Here, initially within the bounds of a sufficiently large and complex project one could build up a coherent system of signification where all program distinctions and social type distinctions and relations are correlated with positional and morphological distinctions and relations. I don’t want to rely on familiar codes here, which would constrain creative invention. Think of postmodernism: the semiotic project was in contradiction with radical innovation and abstraction. Instead we have to start from scratch. How does this work? And can the semiological system have some constants from project to project, or is each project its own monadic system of signification? If it’s large enough it might be okay to stand alone, initially. I have written what you might call a theoretical refoundation of architecture semiology, in Volume 2 of my big book, setting out a series of axioms that structure a coherent semiological project. On this basis I am currently working of semiological design projects with my AA students. In my book I’m referring critically to your work as well, because I realized that your project was not viable as a project of pure syntax. I don’t think its possible to build a syntax without a correlated semantics. There is no “language” if there is only syntax, but you have been rejecting the semantic dimension.
PE: It is impossible. Jacques Derrida told me architecture will always mean. And I have been chasing this elusive goal of asking if it’s possible that architecture will not always mean. Is that possible? If you get to a coded system that introduces a digital semiology that doesn’t exist in a pure linguistic frame, you might be getting something that initially has no meaning. Pure grammar.
PS: I think the problem is, there is no single grammarian who’s ever analyzed any language without intricately knowing that language. If you listen to somebody in a foreign language, you can’t even pick out the units and differences accustically; it’s an amorphous world, and I think methodologically you can’t hold on to these differences you want to keep hold of – a b c 1 2 3 – if you don’t simultaneously build up the semantic order. Even if its not a logical impossibility, it’s a psychological and mnemotechnical impossibility. In my book I paraphrase Wittgenstein roughly like this: Its not a rule if the criterion of following the rule amounts to: whatever Eisenman thinks is the rule the moment he engages in the next step.
PE: You can argue you are involved in knowing the evolution of the history of the urban, but your algorithms are meaning free, the algorithms that generate urban form.
PS: They generate the domain of the signifiers.
PE: Yes, but they don’t have any meaning.
PS: No, but in order for you to discipline and systematize them through grammaticality, through syntax, you need to build the semantic layer at the same time. The semantic layer acts as a control layer, just as the syntactic layer acts as the control layer for the semantic layer. Otherwise you get lost in abstract series - a 1 a b c 1 2 3 c 3 - in no time. Its like double entry book keeping. Without it you cannot reach any complexity without getting hopelessly mired in error and confusion.
PE: They initially are meaning-free.
PS: I found a way in the to integrate the semantic layer, the meaning layer, into the digital design model. I get the meaning layer as another correlated subsystem in my multi-system parametric model. The signifying relation is another correlation within the logic of associative modelling. Specifically, I’m taking agent-based crowd modeling as this meaning layer and program agents to be responsive to designed environmental clues in their behavior; their behavior is modulated by architectural articulation. Any feature of the environment might modulates their behavior, and thus becomes an effective sign or communication. That’s the signifying relation proper for architecture. In the end the meaning of the space is what takes place within it, that’s what it should be communicating. The designated, designed space is a framing communication that invites potential participants to share a certain particular communicative situation. The meaning is the use, the social function. I can bring that social function into the model by crowd modeling and by scripting individual actor’s behavioral rules relative to spatial distinctions. Agents might come into a space and slow down as they move from a marble floor onto a carpet, gather around a central position that they’re invited to gather around by a territorializing ceiling feature. These are not key-frame animations, they are literally programmed agents that move autonomously according to stochastic rules that change in dependency to spatial markers, thresholds, gradients etc. The agents are scripted modulate their behavior relative to selected stimuli, which are the features of the model the designer. So I can say carpet means ‘slow down and orient towards others’ (private places), hard surface means move independently and ignore other agents (public spaces). That’s operationalized, parametric semiology. It generates precisely what we’re looking for: it coordinates the action of free, rule-governed actors, and allows us to test which global social patterns of interaction emerge from which set up of individual rules in combination with which coded spatial frames. This is a methodology for a semiological project you can work on cumulatively now. You can run it and rerun it, recalibrate the signification relation, see if you can generate the kind of communicative situations the client is aiming for. Criteria of success might be: encounter frequency, average encounter durations, range of durations, group formation, average variety of interaction partners etc. All this allows us to maximize the communicative intensity and efficacy of our designs. And how do these spaces look? Complex, dynamic, intricately ordered. So it’s the richness of dense, multi-audience event and interaction scenarios what we should be aiming for in our contemporary network society.
PE: This is very new. We’ve hit upon the core.