The Dean of SCI-Arc Eric Owen Moss interviews Hernan Diaz Alonso and Marcelo Spina, two members of his faculty, about the fluctuating relationship between radical architecture and the tools that give it shape
Marcelo Spina directs his LA-based practice P-A-T-T-E-R-N-S with Georgina Huljich. The duo have won numerous prizes and have built pavilions for MOCA and SCI-Arc. Hernan Diaz Alonso’s office XEFIROTARCH is alson based in LA, and he built a pavilion for PS1 in New York in 2005. Spina and Alonso are part of a new generation of architects who have rejected hand drawing in favour of digital techniques. They are interviewed for the AR by distinguished architect Eric Owen Moss, creator of many buildings in Culver City and elsewhere and dean of SCI-Arc (where they both teach).
Eric Owen Moss: The first question has to do with stages of evolution in the development of design ideas, and the design tools used to implement the ideas. As a primitive example, let’s assume someone said, ‘OK, here’s a rock. Let’s pound these sticks into the ground to hold up the tent.’ And someone eventually responds, ‘This is a pain in the neck. How do we make it easier?’ And the rock tool evolves, over millennia perhaps, to become a hammer. The developmental stage ends. How the rock becomes a hammer? That’s a subject. Once the tool is available, the next group isn’t burdened with tool conception. How do they extend its use? That’s a subject. By extension, in the developmental stage of digital tools, what was the inventor’s itinerary before there was a Rhino menu? When the architect imagines something he doesn’t quite know, with no a priori menu of options, he invents a way to represent and to implement what’s represented or the conception dies. So there’s an interrelationship between conceiving a form, a space, a shape, and inventing the means to deliver the conception. And that job is different from that of the next generation which now owns the tool. The next group doesn’t invent it. Instead, they have to learn it. Let’s talk about this chronology of invention, and how the imagining of an object we don’t yet know, in a time when the tools don’t yet exist, is different from designing and building the identical object years later when the tool is known.
Marcelo Spina: I think there is an obvious difference between when the tool exists and when the tool doesn’t exist. Who produces the object before the tool’s design? Who does it first? When it comes to radical work, there’s a huge difference. I think the parametric capacity, when it comes to producing thousands of drawings that are slightly different but that all share a typical aspect of joinery, for example, is undeniably essential. It changed the construction game. To locate parametrics with respect to the generation of the work, how to represent the development of an idea, when it’s not yet clear how it’s going to be imaged and built − or if it’s going to get built, which may be determined based on what we are able to represent − is a very different discussion.
EOM: Do you have an obligation with the digital tool to extend the meaning or purpose of the tool; so that as each architect handles it, he doesn’t simply pass on what was given to him, but has an obligation to rethink it? I mean, in the same way, somebody could sit here 10, 15 years ago, with a parallel rule and a couple of triangles and produce (or not) a predictable Hitchcock-Johnson redux of a modern architecture vocabulary. The modern language, one could argue, comes directly from the tools used to represent it. Now there are exceptions to that, and that’s what we’re wondering about. Mendelsohn is an exception, or Gaudí. The usual tools and rules don’t account for their work. But, generally, the advocacy has a kinship to the tools that are used to implement it. And if Mendelsohn is sitting in the trenches of Verdun drawing these funny little sketches that don’t depend on a parallel rule and a triangle, is there an analogue digital development stage?
Hernan Diaz Alonso: I think it’s wrong to assume that digital and software tools are stable. The tools are always in flux.
EOM: For you or for everybody?
HDA: For everybody, at least in the ideal sense. I’m not saying that’s always the case, but I would argue that the body of work the people in this room are interested in is always changing and always challenging the limits. The tools change what the limits are. Something that three years ago you couldn’t build and now you can implies a continuing process of improved capacity indefinitely. No reason to assume you will build the same thing 20 years later. The tools are the means to start to think about new ways to challenge what the tool can and cannot do. I think that the discussion with relation to tool capacity has always been there. I mean the idea that some architects are comfortable with the status quo doesn’t preclude a continuing challenge. The other part of the argument is that some of these tools are not only vehicles for production, but what I would call a way of thinking differently about the work. For us digital tools remain fresh and renewable.
EOM: Can you give me some examples? There’s a pivot point in the history of technique which has to do with the representation and related sharing of documents, and engineering, fabrication, construction, all of that. That’s clearly changed. It went from an established way to a disestablished, experimental way …
HDA: … To stable again.
EOM: So are we now re-established with an analogue of the ‘Hitchcock-Johnson’ pro forma?
HDA: Yes. And the hope is we will find new ways to make it unstable again. Especially, for example, if we go outside architecture like in the late 1980s and early ’90s, when the discipline started to flirt with tools coming from movies, animation, video games and so on. Those tools were added to the architect’s repertoire, got re-adapted, then outdated. Software like Rhino came later and it became like the latest applications of a continuing speculation. CATIA too, all those options. Now, if you look at the evolving history of the shape discussion that goes from the beginning of people who were working with simple geometries to plane curves to discontinuous surfaces. Then we move into polygons, then subdivisions of polygons that we can connect. At the beginning we couldn’t connect the pieces. Now you can. Then parametrics and scripting arrives to again diversify for a moment. And next come the robots, and the robots are again destabilising. So, my sense is, yes, digital tools have created a new plateau. Digital tools created a whole new series of agreements in terms of sharing information, which I think has now been clearly established. I would argue that there are certain parameters that are more cultural than technical. You can think about a new philosophical perspective in the Renaissance, for instance. It changed the architectural premise. Another example is the body of work that was triggered in the ’50s and ’60s that I think influenced people like you and Thom Mayne, through axonometric representation. So there is always a historical moment when techniques and methodology influence culture, and the culture argues for a new body of work. What is fundamentally different at this particular moment is that our intellect or the cultural aspect of the practice and the practical applications are much more in sync than they were during the axonometric era. Let’s say Stirling, for example, designed using axonometrics, but he couldn’t build it through axonometrics. Today we have a much more direct correlation between digital representation and construction implementation, and I think that is a fundamental change.
EOM: I don’t know whether the explosion of digital capacity actually has narrowed the intellectual perspective on architecture and focused attention on a relatively narrow venue − engineering, fabrication, and construction capacity. Is there a correlation between a new representational capacity and what you refer to as a new intellectual/cultural point of view? Are we stuffing into odd shapes what we used to stuff into even shapes? Or is the programmatic content evolving too?
MS: The problem for me is that you have the technical means, software means. You have the calculus to allow others to share and understand content − engineers, contractors, fabricators. But it’s still pretty difficult. Second, not everyone is interested in doing the same things. So, I think the problem of the relation to content is very much at stake in relation to differing individual points of view. It takes a while to kind of get acclimatised to a certain set of tools and that, at least initially, creates a myopic relation to broader content. At first the tools have to be mastered. Then you can begin to discuss new content and history.
EOM: Let’s say 60 years ago on Park Avenue, Mies puts up a building, and the argument for the programmatic content of the building is essentially orthogonal, gridded space − open, flexible, changeable, malleable. The theoretical argument is that any undue spatial complexity would compromise flexibility over time. You don’t like the space or its subdivisions, you move them; you want it open, people connected, OK; you want it closed, OK.
MS: For many people that concept still holds today. That’s not a battle that has by any means been won or lost.
EOM: Now let’s take a digitally invented tower in Singapore. The elevations dance. Mies is gone. Let’s say the elevation of the tower is kinetic, changes from hour to hour. We’ve built a clock/calendar that responds very precisely to its environment. The case for variability is the opposite of Mies’s flexibility argument. Now, instead of right-angled, gridded neutrality, flexibility means the opposite, an idiosyncratic shape. Now the argument is that digitally enabled specificity negates the requirement for future flexibility. So shape variability, it’s argued, then allows you to accommodate a range of different uses and sizes, different apartment types, for instance, as the tower undulates in and out. What happens if everybody wants a two-bedroom? The Mies argument was perhaps disingenuous. It really had to do with a predilection to representational and constructional tools and methods. The Singapore Tower image is also a conceptually disingenuous argument. The digital tool allows you to make shape and space, which we rationalise in terms of a plausible, flexible human purpose. I’m not sure you can trust either one of those arguments.
MS: I know it’s important from the pedagogical point of view to discuss these things, but I always associate the answers with a personal, core set of beliefs on what architecture should be. I don’t think we should all necessarily do different work as a consequence simply of the availability of different tools. You probably will build differently. You probably will communicate differently. You’ll probably work longer hours because there are more options. But I don’t think the relationship of tools to design content can be generalised. I have a problem with the argument for greater efficiency or higher performance levels strictly as a consequence of new tools. I don’t agree that a concept should exist simply because the tool allows it. The architect is delivering on a set of core beliefs. The new image should belong to the a priori belief system that precedes the enabling capacity of the tools, not that follows the enabling capacity. Unfortunately the latter is often the image of contemporary work: that it’s simply derived from the new tools.
EOM: I’m asking if the content drives the tool somewhere the tool wasn’t intended to go? Or, more disingenuously, even though you claim the content is driving the tool, isn’t the tool surreptitiously driving the content?
HDA: I don’t agree with Marcelo’s assessment. I think there is a direct correlation between the tool and the final outcome. I don’t think it’s disingenuous. My point is, if you think about music, the sound of the Beatles or Jimi Hendrix would not exist without electric guitars, amplifiers and wah-wah pedals. The musicians would be great. They would have done other music. The aesthetic sensibility would be different without those tools. So the tools at any particular time enhance, change or define the content of that time. So if we are sophisticated, we try to merge or create a friction between our individual desires and whatever tools are on hand. I think you can define tendencies in architecture and art based on the representational tools available at the time. If you take the example of Patrik and Zaha, I would argue there is a radical difference between the work of Zaha Hadid as a painter, Malevich style, and the introduction of the computer and Parametricism. It’s not the same. If you take the Vitra Fire Station in Weil am Rhein, the project was based on fracture and a clearly different sensibility delivered by the new tools.
MS: It’s true that tools evolve and the work evolves. But it’s also true that we all use the same tools but produce very different work.
HDA: I would say that the work I produce today is heavily driven by the tools. If I were not using these tools, my work would be different.
MS: If you’re going to try to tease out the subtle relationship between tools and one’s own work it’s an intersection between core beliefs and tools. Sometimes core beliefs change as a consequence of tools.
HDA: There is a down side. There is also a McDonald’s version of your work − a version any mediocre corporate architect can reproduce. That’s a problem. Then the tools do take over, and the tool is identified with certain formal preconceptions. I would argue the same is true with music and film.
EOM: It’s interesting that the tool can be understood as a means to efficiency as well as invention. To be fair, the new tool has an interesting neutrality which facilitates, in some hands, a cheapening of the work − even more austere, even more efficient, and even more penurious. You can go back today, for instance, and rebuild the Seagram Building using Revit.
HDA: Yes, and many do just that.
EOM: And AutoCAD too can be an efficiency tool. You can see the variability of production options in terms of how the tool is applied. And what we are supportive of is an inventive minority and how they apply the new tools, as opposed to how they’re generally applied in architecture and engineering.
HDA: I agree, but I will say redundancy and repetition occur with or without the tool discussion. Item one: Mies makes the Seagram; item two: 40 years of boxes from SOM. They’re not the same thing. One thing was Seagram and Mies and conceptual ‘neutrality’ as a design goal; it’s another thing to take that design hypothesis and make it into a style or an efficiency type. The problem remains. It’s a fundamental problem in architecture.
EOM: Mies’s box could be explained as an early 20th-century reaction to, for instance, the Neo-Classicism of McKim, Mead & White. To make a box, let’s say in 1915 or 1920, was a search for a new archetype for public architecture in America. It was looking to replace the Boston Public Library − McKim, Mead & White, which had a big reputation in Boston, the centre of American culture. To make the box 100 years later when the adversarial relationship between McKim and Mies is long gone doesn’t mean a thing. That conflict between old and new is gone. Another related topic is what I would call a radical idea versus the image of a
radical idea. Whether something is genuinely radical is so only if it hasn’t been done, or is difficult to do, or the results are uncertain, which is different from reproducing years later what the profession learned to do in former times. The struggle to implement and to understand new work, as a valued act of architectural conduct is very different from repeating a learned image. Now we notice many buildings that are carrying forward the image of what’s supposed to be radical, but it’s now a learned response as opposed to a discovered response. And the question is whether the digital process facilitates this ‘living up to an image’. An architect makes recognisable buildings. At an early certain point, one doesn’t know how to do it; at another point, one does. At an originating point, it’s a radical act;
at a later point, it’s actually conservative.
MS: I think the challenge for me today in terms of representational tools and techniques is how to customise them. I want to make them more personal, make them more akin to the premise of my own work, less generic. It’s not just the software. It’s using a combination of a tool set in a very particular way to produce very particular forms or very particular conditions that belong only to a single architect.
EOM: So you think the new representational tools allow you more opportunities to customise a point of view?
MS: If you use them the right way.
EOM: What would be the right way?
MS: The right way means your own way. The way we model surfaces in Maya and then translate them to Rhino is very different from many of our contemporaries.
EOM: Do you have an example of a particular project?
MS: For the SCI-Arc Pavilion, there’s a contrast between two shapes that aren’t typically found together − we are very interested in creating that sense of disjunction between one side, completely curved, and the Cartesian condition on the other side. One is more hard edged and faces the parking lot; the other, softer side faces the street.
EOM: And if someone were to come along in 50 years and look at that and look at the pavilion, knowing historically, if you think like X, you do X, if you think like Y, you do Y, but Marcelo came along and put X and Y together in an unorthodox combination.
MS: Let’s say that because the software now facilitates complex curves, which were once difficult to produce, the assumption now is everything is supposed to be curved. Same mindset once applied to boxes. Neither helps.
EOM: And what, in the broad human perspective which representation delivers, what would be the conclusion about that combination of topics?
MS: It means for me there is a certain engagement with conventions, that it contests the regulations, personalises the work.
EOM: Well, if you look at the movement of architectural ideas over time, they don’t necessarily get more radical or they don’t necessarily progress in a line to a new design destination by undoing the previous idea. I’ll give you an example. You go to Paris, the Eiffel Tower and the Sacré-Coeur, let’s say: two examples that could be two opposing ways of discussing the meaning of buildings. The two were done at roughly the same time. One is somewhere between Roman and Byzantine. The other is more an act of progressive engineering. The new engineering design implies a revised perspective on everything else in the city, makes everything in the city different in contrast to the new tower, because, in a way, if the tower is right, everything else has to be re-examined, looked at differently. So there’s also an essential stabilising Sacré-Coeur conservatism, across time. So re radicalising content using the digital tools we’re talking about could also be used in the service of very different ideas.
HDA: I still believe in the idea of the radical project. Radicality has never relied completely on the means to deliver it. It also requires a relationship to society, to culture, at any given time. The notion of time is tricky. When I started working with computers in around ‘98, I completely embraced a radical change in the way I produced or thought about architecture. I was a pretty gifted drawer; I knew how to do that and I felt very comfortable about working that way. When I started with the computer, I decided to embrace it and completely abandoned any vestige of my previous life of representation.
EOM: Do you draw anything any more?
HDA: I draw almost nothing.
EOM: So with students, do you ever draw?
HDA: Most of the time I grab the mouse and do something directly in the computer. It was a fundamentalist’s decision let’s say. You go to my desk in my office. There’s nothing. Only my laptop. That’s it. There’s nothing else. My way of working with the software was an embrace. The parametric software that I use is animation software. The parametric software in my office is not Rhino, it is Maya. We use Rhino to translate Maya into a construction format. I was very interested in the cinematic aspect of software, and we have been using the animation on the camera as a way to understand architecture, the way to articulate and produce a cinematic aspect. That’s why I feel so comfortable working with the computer, not only because the tools were cool and allowed us to do other things which is in itself interesting, but it was because it really was a way to marry my original ambition to the new tools. I realised that the work I was doing in school in Argentina − drawing by hand − and what I’m doing now, at the core, are still the same. I think as an architect you’re always working on the same problem, but the techniques and the tools shape and change, and I think we establish a certain level of partnership with the new tools.
EOM: We’re talking about a kinetic tool. If you’re looking at animation software, what moves, conventionally, are the people who are in it, or see it, or fly over it, or pass by it. But the building pieces are fixed. So to understand the spaces or relationships of spaces you’re arguing that a kinetic mechanism gives you something different from a static mechanism. But what does a kinetic mechanism allow you to understand about building?
HDA: It’s what I would call the fourth dimension. It’s to understand space or an object as an entity on its own terms. So animation is not just a mechanism to represent the work; it’s a way to generate and to think about the work in the same way that somebody else like Frank [Gehry] will work with models, or Thom [Mayne] will work with axonometrics, or you with all those sketches. It’s a parts-whole relationship that one needs to establish. For me there is a level of distance that I need to establish from the work. The camera and the animation allow me to do that. Really, it’s a pure act of object autonomy, almost like the object doesn’t belong to me, which is not true. But there is something about that kind of schizophrenia − I own it; it owns itself − that’s important.
EOM: Let’s say you’re making something, and, in the course of making it, you adjust it. It evolves. It becomes something else. How do you use animation software as a way of thinking, to make judgements about form and space and material? That’s a very unconventional way of thinking because you introduce movement. The CIAM guys always talked about movement as routes of circulation. You look at the drawings of Chandigarh − seven velocities, seven circulation forms. But that’s not what you’re discussing. The velocity theory − animation design − raises a number of questions: the rate of the animation, the perspective of the animation?
HDA: What is the camera? What is the angle?
EOM: What is the speed?
HDA: That is part of the problem. The other one is how you start to work. I think also on ways to introduce limitations on problems. I’m not saying architecture is movie making. For instance, we’re doing a centre for Boeing in Seattle, which is an interior scheme. I felt the process of designing the project was very similar to drawing by hand. It was done through animation and image agreement − understanding of the nature of the space and the form. Now, when we’re producing the documents in order to build it, we’re using Rhino and CATIA, as many offices do. I would say that popular software and the related technology interests me the least because I see very little angle for corruption there. It becomes a very straightforward part of the project. So I’m not saying that I’m drawing. I was saying that when the team shows me a whole DD set and you look at it, you will make a mark; change this or change that. You don’t go back to the computer because those already are fixed drawings. That’s it for hand drawing. I would not consider CATIA and Rhino software tools for thinking. They’re for production. AutoCAD is simply a more efficient way to draw by hand. Even when I was working by hand in the office, I’d never be the guy who did freehand sketches. I always worked with triangles and compass. Everything is straight, very precisely drawn. Even what I would consider to be a sketch was an incredibly precise geometrical drawing. I never did a freehand sketch. I never trusted them. I’ve never been interested in that aspect. I’ve always been obsessed with precision, and, for me, the computer was that on steroids, almost a detachment from the sense of the hand.
EOM: So you could explain how the animation theory of conceptual development might differ or might give you a certain result with more conventional tools. This is a different idea.
HDA: In my office we tried every new tool that came out. We rejected many of them because there is a certain essential conservatism that emanates from the computer world.
EOM: I have a doctor like that. Every time somebody delivers him a new pill, he tries it.
HDA: I always go back to Maya. So there is a certain − even in my own way of working − a certain established software premise in the way I work because I’ve become comfortable with it and I understand how to work with it.
EOM: This is always interesting of course. However radical the work is …
HDA: … There’s always a conservative aspect …
EOM: … Something you can turn your back on and not get stabbed; something you trust.
HDA: In that sense it’s no different from you with your guys and your team. In Maya, with my team, there are guys who are better than I am right now with Maya. They know all the scripting which I am not familiar with. So, even in that, I have a relationship to my team, similar to the one you have with yours. You do it through a series of sketches, they go back to Maya, then translate it into a more sophisticated software. So that similarity in use of software, regardless of different points of view, never goes away. Something essential to all our processes. I agree with Marcelo − we have to have a very clear conceptual apparatus, a chronology of softwareapplications, while you’re trying to produce architecture today.
MS: One, let’s not get too dogmatic with respect to tools because that actually freezes the work. But there has to be an inner coherence, which is where work differs. I wouldn’t want to say signature. How about …
MS: Trajectory? Value? The way I used to sketch when I was at school was very different. I would tend to sketch certain objects in plane and others in perspective; you know, try to study things in sketching. We had just finished a competition for a mosque and were starting the pairing and facade using Maya. Then after that, there’s already probably a script in Rhino and Grasshopper that produces what’s needed mathematically because I believe that however subjective and random is the solution, there is a drawn, mathematical solution. I’m thinking all the way to a final object whether it gets built or not.
EOM: But what you’re saying is a free form has an underlying technical requirement for implementation. Even the most complicated fractal shape can be reduced to a mathematical conception. The freest, most intuitive shape, to be constructed requires an antithetical spirit. Enormous rigour. The computer might or might not give you the first; but it certainly guarantees the second. That’s why the odd instinct is built more often now. The computer rationalises the instinct, makes it technically plausible.
HDA: For sure.
EOM: You used the word ‘corruption’, but you never explained why corruption is an aspiration or a virtue. What is corruption? In the first instance a software designer designed the software anticipating it would be used to solve certain problems in a certain way.
HDA: Exactly. It’s a system. It’s a method. And I would argue that anybody who produces a work of interest with software, as a by-product, has to find different mechanisms to corrupt that methodical linearity that underlies the software design. Corrupting means finding possibilities to create an accident, possibilities to contradict; to prove anomalies within the frame of reference of that tool’s use is actually possible. I can argue that the corruption tendency you can see clearly in the last five or six years, was much more difficult to discover in the first 15 years of this process. Everything’s a circle and the opposites almost touch. Now, we have tools and techniques that are mathematically perfect. You draw a curve there, it’s perfect, mathematically, and naturally, there is a desire to find agents to produce the opposite, to introduce imperfection.
EOM: The first step was the advent of software that allowed you to conceive in a precise way, what was very difficult to represent before the software. But now the software is available, the goal is not to deliver what the software initially allowed, rather to find a way to contradict its original intention.
HDA: Yes. The ultimate goal of the software is to enhance the humanism of architecture.
EOM: What do you mean by that?
HDA: You have more conceptual options, more and different ways to represent the world. If you’re not willing to explore that, if you’re not willing to challenge the known form language, you’re simply serving the purpose of the original invention.
MS: I think that’s actually important that we look for materials or assembly conditions that have nothing to do with the software. Departures from the original menus create a striation, what I will call a certain authenticity in the work, which I might call corruption. I like that word less, but I think there is certain alignment in terms of the continuing goal to produce new work.
HDA: Somebody should write a book about which software succeeded and which failed. Remember the middle to the end of the ’90s, formZ was super powerful and then it disappeared. Why? It was based on fractals and was abandoned and replaced by something else. Perhaps not enough time has passed to draw conclusions on what worked − or was corruptible − and what didn’t.
EOM: The failure of the incorruptible.