bookshelf      info       instagram

Old School

Jesse Reiser & Nanako Umemoto

Jesse Reiser and Nanako Umemoto speak with inter·punct about the role of technology and theory in architecture.

For additional information and to view Jesse and Nanako’s full lecture, visit the CMU School of Architecture website.

i·p: The inter·punct project is motivated in part by the idea of creating an inventory of the state of architecture discourse today. We want to ascertain what it is that people talk about, but also what’s missing from the conversation. Beyond the fact that we’re not discussing projects in the same way we used to, what do you think the state of the conversation is today?

Jesse Reiser:
I think there’s a lot of conversation—a lot more discourse than ever before, but it’s more diffuse. There are no longer clearly defined movements that have priority over others. There are a lot of different discourses happening all at once, none of which seem to be dominant. Personally, we’re in our own world. But, it seems that when you survey the scene, there’s no lack of discourse. Rather, there’s just so much that there isn’t a clear hierarchy.

i·p: So if the conversation is filled with more chatter, how do you situate yourselves within that?

We’re the hermetic ones. [laughs]

Nanako Umemoto: We’re just always independent.

JR: I think we have our own interests and so we try to seek out opinions in a direct way. There’s enough momentum within our own work that, from the outside, we may seem to be carrying out projects that could be exhausted, but from the inside we see the problems differently—for better or for worse.

i·p: During your lecture at Carnegie Mellon, you mentioned that you like to learn from competitions, or rather, that you learn from your competitors, specifically. Can you elaborate on that?

That was in reference to the scene at Columbia in the 1990s. As junior faculty, we would all enter the same competitions. Stan Allen would enter, we would enter, Greg Lynn, Hani Rashid would also occasionally enter. None of us won. [laughs] But, afterwards there would be a sharing. Because it was the same program, there was more often than not a sense of similarity rather than difference among the competitors. We weren’t talking about radically different projects, but were trying to get at similar issues. Greg Lynn would try to sketch everybody’s scheme before they showed it to him, to guess what they would do. It was funny.

NU: It was not to laugh at other offices though.

JR: No, it was more that there were common interests. We were discussing what not to do, as much as what to do.

i·p: Do you think that kind of debate still plays out now that everybody’s in practice? Do you still compare notes with those people you worked with at Columbia?

No. It evaporated. It’s long gone. We were junior faculty then—none of us were that established.

NU: And nobody’s there at Columbia now.

JR: We don’t talk about the work so much now, but we have a history with one another. Maybe that’s the natural course of things, like rock bands.

i·p: What role does publishing play in sharpening your own polemics, versus your peers? How does publishing change how the problem looks from inside?

The Atlas of Novel Tectonics was a distillation of what we found out in our teaching. A lot of the points there are veiled criticisms of fellow practitioners and projects. In truth, a lot of the material came from presentations and arguments based on specific issues we were trying to work with and later generalize.

i·p: Usually things like the Atlas are interpreted as programs or manifestoes, but, as you said, it’s built out of criticisms of other projects or maybe other figures or ideas. How does that critical function motivate your work as an architect?

A large section of the book is a criticism of the state of the surface project and of the ideas circulating at Columbia at the time. We were trying to bring in case studies of tectonic systems rather than move into detail from abstract notions of surface. We were looking at concrete construction systems either within architecture or engineering that would productively criticize the ideal notion of surface—the infinitely thin, featureless architecture that came out of computation. We used case studies rather than a purely argument-based or idealized notion of what surface would mean.

i·p: For many architects the limit of the ideal mathematical surface is found in the process of getting something built. I want to ask about O-14 because the Atlas predates the tower just barely. Could you talk about how the ideas and arguments of the Atlas were followed-up in your experience designing O-14?

O-14 had to confront in a real way a lot of the issues that were introduced in the Atlas. There was a bit of a paradox there, where we realized that there was a lot of argumentation against representation in the Atlas, a legacy of the 1990s. We all were trying not to be semiotic. We were trying not to deal with the obvious historical references, but nevertheless in moving through an actual project you realize their significations are unavoidable. So, rather than rejecting them, I think part of the work on O-14 was to combine the more literal, physical, tectonic interests with the effects of the building.

i·p: It’s very interesting that the space in which you confronted the semiotics of the building was not the theoretical space of the speculative project, but rather the real process of construction.

Right. We were ourselves shocked to see that when the thing was finally built it looked like a built rendering. We probably should have known this, but it was a kind of signification that was really shocking to us.

i·p: In 2013, at the In Pursuit of Architecture conference hosted by Log at the Museum of Modern Art, that was the discussion around O-14; the building in some ways looked like a built rendering entered into this strange signifying relationship with the rendering.

In the urban context, it became even more strange.This relationship between the building and the representation, in which representation functions as a two-way street, was something that I remember having a very strange exchange with Steven Holl at Columbia about. He got very upset with me because I had an intuition that there really wasn’t a gap with representation. In 1950s-era wash renderings of office parks you get exactly that. The trees start looking like representational, the sky starts looking representational, those representations then fold back into the world in a strange way.

i·p: But that relationship wasn’t something that you had anticipated.

No. I had that intuition about the render and he hated that argument. From a phenomenological point of view it disturbed him, but then it came back even more with the technology in O-14.

i·p: What role did technology play in your design methods and the construction of O-14?

Well I would first say that while our office certainly uses computers for almost everything, we were trained a generation before computers emerged in widespread use. I think there’s always been a productive tension generated between what the programs want to do—the inertia of the programs—and what we want. Certain architects celebrate this, but for us it’s really a painful process. We’ve never really worked purely with the computer. We mix analog techniques with the modeling software—we don’t trust anything on the screen. In the perspective window most of the evaluations are essentially two dimensional.

i·p: So, for you, the ideal medium of evaluation is the model, the object.

The physical object is the final arbiter. I believe it. What’s in the perspective window is always distorted. If it stays in the screen then it’s fine, it’s in its own world and it can have its own reality. But, if the aim is the artifact, then we go to the model to check.

i·p: In terms of O-14, what were some of the things that you were evaluating as you were designing?

We were working with a guy who was really quite an expert at scripting, and he started to script the holes, and we were never satisfied with the products of the scripts. The holes appeared too mechanical and so we were always messing with them. We actually started with a larger field of holes and tried to script the entire field.

Then it came down to us adjusting the gradients to such a degree that our script-writer would have to try to re-write the codes to fit our adjustments. It all became so localized that it was absurd. Every situation was handled differently, and so, we decided on another way of dealing with the apertures. We went with a graphical approach—and Photoshopped over it.

i·p: To what degree are the holes structurally determined?

There were parameters set up by the engineer according to the scale of the hole and how much material would be used, but there was a tremendous amount of flexibility within those terms. Each hole could drift within the grid and be rescaled. The size and position of the holes changes the strength of the concrete, too. All of these different parameters were more or less objective, but when it came down to it, the holes appealed to an aesthetic side of design.

i·p: So in the end the structure was essentially determined graphically.

Much of it was graphically determined. There were still functional parameters but much of it was aesthetically tuned, which was very disappointing to the scripter. He was annoyed because, to him, the changes created randomness—it wasn’t systematic. He could only see what we had done as a form of randomness, but we weren’t working with randomness. The denser areas of structure had to drift toward a column grid in the basement. So, part of our work was trying to defeat an obvious read of force flowing down the building into the regular pattern of the column grid. Our challenge was a game of working against the regularity of that system. Our engineer called it a “trickle-down” logic.

i·p: It’s interesting to hear you say that because it suggests that the project was working with the signifier-signified relationship even as you intervened on the rational design of the structure. The representational project is co-incident with a structural one.

It was the same with the bridge for Alishan. We sent our model to Arup’s Advanced Geometry Unit and they put it through an optimizing process. But I hated what came back. The dynamic is also about architectural will versus a purely objective, optimization logic of engineering. We’re always in tension with that too.

i·p: What do you think is the motivation behind defeating the easy read, or the straight-forward read, in which the object communicates the flow of force?

In that case I think there were just other agendas related to the aesthetics of the bridge. We didn’t want it to look like an engineering project. Brett Steele made a comment about the project that I thought was really beautiful: there are certain engineers or architects who want to find a way of getting the forces out of the building in the most efficient way possible. I think part of our interest is slowing that process down or complicating it. It comes from the aesthetics of laminar and turbulent flows, but it plays out programmatically, too. Those effects would also fall into line with the ways in which we organize space or organize program.

i·p: I want to shift the conversation toward the status of theory in your work. In your lecture you said something quite provocative—that you “design first, theorize afterward.” This challenges the traditional claim to theory as an idea that directs how design ought to be pursued.

We see theory as part of a reciprocal chain. Of course we have certain assumptions when we start a project which come, in part, from previous experience. It’s not as if we’re not thinking before we start, but we try to see the design process as something that will produce a project we could not have thought of beforehand. We already know this from experience, and even if you have a concept, it’s always different from what you end up with. At a certain point, the design starts taking over, and either you’re going to fight to try and get that design to become more like the concept—this would be our position—or you suddenly see that either the initial working concept is not viable, or that the new information could be integrated. The experience becomes part of the way you theorize the work. Otherwise, why do it?

NU: We have been working this way for the last three decades. Every time we start, we start from where we were last time. We always have some kind of base to start from, but we can’t guess what will be produced.

JR: Or, you’re arriving at something that was latent and waiting to happen that you might have started fifteen years ago.

i·p: Could you tell me a bit about your experiences over the past few years putting together your monograph? Did you find it challenging to open your archive?

We know what kind of projects we’ve done, but essentially we had to theorize and really gather all of the information, images—everything, together.

JR: I think one of the things that came out of it immediately given the way we work is that we couldn’t put this thing together by decades. It immediately seemed like it would be much better to find thematic threads across periods of time and to connect those projects.

i·p: It seems that for many architects the process of assembling the monograph is a challenge of narrativizing both the practice and the body of work. Were you surprised by the readings of your work you had versus those that your peers brought to the monograph?

Part of the idea of this is to bring in essays or parts of writings by contemporaries at the time, either on the work or comparing the works with others. The book is interspersed with that material.

i·p: Are there projects that are anomalous with respect to the structure of the book, projects that don’t fall cleanly within the thematic divisions?

Some projects more or less stand alone. Some small booklets we’ve included are speculations, separate from building projects. Like the Architecture for Dogs project.

i·p: Obviously in your practical work you have interests and agendas that you pursue but how do you work with these in your teaching?

We share often the program with our students. But I was teaching with Jesse in the 1990s for a few years, and after he moved to Princeton I stopped teaching and focused on the practice. Since 2009 I’ve started teaching again.

JR: I was recently talking with Guy Nordenson about this—he has a very different idea of the role of a teacher than ours. For us, it’s important to work on the problems we’re actually working on in practice. Not that we’re going to have students work out projects for us, but that the themes, especially in the studio work, would be of immediate of concern for us rather than separating pedagogy from the practice. I think we’re really trying to work through problems that are suggested by the work we’re doing. It becomes an opportunity to get feedback. We’re throwing these ideas out and we want to be surprised by what the students’ reactions are.

NU: But the way Jesse teaches and the way I teach are very different. The result of the student work is really different.

i·p: It’s another challenge to your thinking and practice.

It’s irritating, even painful at times.

NU: If you have twelve students you are doing twelve projects, because you have to address each student’s interest, and then you can develop that together. It’s very interesting.

JR: As a student I was in studios that were entirely of a different model. With Aldo Rossi, when you were in his studio, you did Rossi. And to a certain extent John Hejduk was like that. But Rossi was doctrinaire. But in retrospect, it wasn’t so bad. You internalize that way of working, for better or for worse, and then you have to work your way out of it. If you accept it, you admire it—and then you’ve got to work your way out of it. Once you realize you can’t do it better than him then there’s a crisis moment! [laughs]

NU: But we don’t push them to do our work. Maybe Jesse does, I don’t. [laughs]

i·p: It’s interesting to hear a defense of that kind of pedagogy because it’s so often derided now.

There’s some value to it... but as a student you also have to be attracted to it.

i·p: Or else it’s really miserable.

I was attracted to Rossi’s work. It was absolutely compelling. Then there comes a point when you realize that if you do this, and even if you do it well it’s not going to be as good as his, so then how do you work?

i·p: It seems like a challenge to empathize with that mindset.

It’s a different way of working. It’s not, “I have this concept, I’m going to dismiss this line of work through a logical set of arguments.” This is more a question of how you would work yourself out of the problem by working in Rossi’s method, or seeing how far you can push Hejduk. I think that’s also what happened with us. In the early work there were very strong figures of Hejduk and Rossi who absolutely had an influence. Hejduk’s tangential touch problem is still present even in the Cardiff [Bay Opera House] project.

So I think that also is a very important point in terms of how we approach a project. There’s an evolutionary model, and there’s also a model which is based upon argumentation, assembling arguments and making decisions before you even get involved. It’s an ideological approach.

I think there’s a lot of conversation… but it’s more diffuse.... There are a lot of different discourses happening all at once, none of which seem to be dominant. Personally, we’re in our own world. But, it seems that when you survey the scene, there’s no lack of discourse.