Interviewed in Belgrade 18 March 2011
Japanese architecture after the Second World War is one of epicentres of modern architectural culture, characterized by an overlapping of international tendencies and local peculiarities. The evolution of Japanese architectural thinking certainly does not rely merely on the fascinating tradition of open spatial concept that inspired Western Modernism as well, but is also a matter of an entirely specific milieu continuously generating vital and new architectural concepts. These concepts are equally a consequence of the rapid processes of modernization, as well as of an attitude which observes modernization from a critical point of view. Sou Fujimoto is one of the leading representatives of the new generation of Japanese architects and his research is based on bringing together an empathic relationship with the contextual given and radical questioning of conventions. Unusual and motivating contradictions between super-modern tendencies and a kind of ‘naivety’ that characterize Fujimoto’s work attempt to establish a new relationship between the physical space of corporeality and experience of the space of a dematerialized networked world.
ORIS: The title of your first book is Primitive Future. At first glance, it’s a contradictory term. Can you explain the concept of ‘primitive future’? Is it a critical message for contemporary society and culture in general?
Sou Fujimoto: Yes, I agree, ‘primitive’ and ‘future’ are contradictory. Every architect thinks about the future, but the future is not a ‘future-like’ future. In the future, architecture will be used by people. I like to start from this basic, fundamental point. For me, people are somehow primitive, because we have an animal-like body and instincts. Of course, we are not animals, but our behaviour, the relationship between our space and body can be very primitive. We go back to that point and we ask: what is architecture, what is a space, or what is a place for people to live in? It is not a critical attitude to contemporary conditions, I just like to start again. Many architectural things changed at the beginning of the 20th century, but after one hundred years, I think it’s time to think again about the fundamentals of architecture. At the beginning of the 20th century, we didn’t have any concept of IT technology or ecology. We can include such new concepts and innovate a new but primitive phase of architecture.
ORIS: You started your practice by introducing spatial concepts which relate architecture and the body in a way different to conventional functional architecture. These investigations are exemplified in the Primitive Future House project which became a sort of prototype for your work.
Sou Fujimoto: About ten years ago, I didn’t have any real project, I just started my office, so I was thinking that I would like to create some kind of a starting point or prototype, like Le Corbusier’s Domino system. I was thinking that I would like to go back to the very fundamental relationship between space and the body, to rethink what a floor is or what furniture or a room is. Then I tried to be free from this kind of existing context. The idea is very simple: combine architecture and furniture together. We can sit on 35-centimetre-high steps, we can use the double-sized 70-centimetre ones as a table. After the first inspiration, I was thinking more deeply to establish such a fundamental concept of the primitive future.
ORIS: It is simple, but it is also radical because you combined traditional elements of a dwelling and blended them together. In your prototypical project and in the Final Wooden House a step is a functional place to sit or lie down but also an element for free-form spatial articulation.
Sou Fujimoto: From the very beginning I was interested in simplicity and complexity. The first inspiration is very simple, but then I found that such kind of a landscape or space could give us creative possibilities to behave according to various situations. After three or four years, I was thinking about the contrast, or comparison of a cave and a nest. The nest is a well prepared space, like 20th century functional architecture. A cave is not a prepared space, but it is a landscape for people to behave or react in. Both the cave and the nest are the very beginning of architecture, but they are very different. And then there is a space full of inspiration, full of possibilities for interpretation or reaction. It can provide a richer experience for people. Of course, I just made the design of the space, and afterwards I was thinking about what it is, and I got a new interpretation, or new concept from that project.
ORIS: Primitive Future House is more related to the concept of a cave, because the cave is less predetermined and you have to discover its potentials. When it comes to the nest, it is predefined in terms of function. You are offering an open situation where people should discover by themselves how to colonize the space.
Sou Fujimoto: Yes, this is a big difference. If we prepare or make a completely open space, it’s a kind of 20th-century flexibility, but I think it’s not so flexible. In the end, it’s just a big space, and people will be embarrassed how to use it. Primitive Future House is not like a universal open space, but more landscape-like, full of many different areas.
ORIS: Lebbeus Woods said, I will loosely paraphrase him, that the size of the space and its performance or function are not directly related, especially regarding housing, criticizing orthodox functionalist thinking.
Sou Fujimoto: The first inspiration for the mini house project was partly based on the typical situation Japanese people live in. We don’t have enough space so we have to be three-dimensional. Both the practical situation and the conceptual situation and my personal impression about the ideal space for people, such a mixture created the concept.
ORIS: This concept also reminds me of Yona Friedman’s investigations of open grids on top of the city which are completely flexible and ready to be colonized in many different ways which cannot be anticipated by the architects themselves. When we speak about flexibility, at least here in the West, we tend to refer to the Japanese idea of flexibility and multi-purpose spaces.
Sou Fujimoto: Japanese traditional architecture is in a sense a very open space. It has many thin wooden columns and sliding doors. We can control the size of the rooms, we can control how the inside and outside will be divided, so in that sense it is flexible. At the same time, I think that I’m more interested in another flexibility. People cannot control the space but can find a way how to be more creative to use and react to the space. I think that in my creative attitude Japanese traditional flexibility and some kind of new flexibility exist together.
ORIS: The spaces you create are fundamentally non-hierarchical. You often deploy overlapping of plans, for example, in family houses where you try to stimulate social interactions by different openings and relations between spaces.
Sou Fujimoto: Non-hierarchical space is one of the most interesting points for me, because it could be related to IT technology, but at the same time simple non-hierarchical things are not so interesting. I like to create more gradient spaces, a dense space and less dense space. People can select a more dark space or open space, a kind of a landscape-like situation. The real space is not the same as the Internet, so I like to translate Internet or web-like situations to real space. I think that a forest-like space is much more interesting. A forest has many trees, it has a non-hierarchy, but it has different areas, various different spaces. A non-hierarchy, but not entirely equal.
ORIS: As soon as you create relationships you get some sort of system which can be more fixed or more open. You introduced some sort of loose complexity in the Children’s Hospital in Hokkaido.
Sou Fujimoto: That was one of the first attempts to create a kind of systemless system, formless form. In that case we used blocks or cells to create random layouts.
ORIS: You try to open up multiple ways for people to inhabit your spaces, but you don’t work with physical changes of the spaces. Instead, you work with various constraints. It’s a completely different way of freedom of use, of making a choice within a physically fixed space. We can learn to accept the constraints and make them work for us, but you have to engage yourself with the space on a different level.
Sou Fujimoto: I totally agree because finally architecture has very huge constraints for me, even though we can move sliding doors, the whole area is strictly defined by the architectural system. This sliding door flexibility is a very small part of architecture, so I don’t want to cheat people. Architecture is not flexible, I like to start from that point. It is very strong and we couldn’t change such a very fundamental point. If you are flexible in your mind, then you can behave according to the situation, or according to how many people you are with. It’s a kind of drastic change of attitude. Flexibility is not about movable things, flexibility is about relationships between the space and your body.
ORIS: Whether we like it or not, architecture always determines our existence in space. Architecture is a physical fact, so there is always the imposition of concept on the inhabitants of architecture. How do you deal with this issue when you design family houses, for example? In a way, you orchestrate everyday family life which is a sensitive and complex issue.
Sou Fujimoto: I face a big contradiction in my design process. I like to design some kind of non-design things, but I have to design, I have to define. I think I create a kind of very simple form or system. For example, I have designed a box in a box in a box house, with many openings, a very simple concept. But how to arrange the position of the openings, or how to arrange the plan depends on the surroundings or how many people there are, or which direction the house is facing. With this kind of more practical planning or design we have to think about reality, and then the concept and reality can exist happily together.
ORIS: You start with a very simple idea and concept, and your working process unfolds through many iterations and tests, mostly through models until you come to the final result.
Sou Fujimoto: Of course, finally we have to fix the exact position of the openings or something, and that process for me is very mysterious. We make the first model which has very rough layout. We then imagine what is happening inside or what is happening throughout the house, and then we start adjusting. I don’t know what the definite end of that design is. It grows like a baby. Sometimes it grows towards bad things, sometimes towards good things, so I need to change its direction. It is not an open and endless process, but it comes to a certain goal. We discuss it with the client and the branch keeps growing, but we don’t know which direction is the best. And then, at some point, determined by the schedule or determined by other reasons, it comes to an end. There is no best direction, so it is very similar to the evolution of the species. I think that human beings, or for that matter dogs or birds, at the time being are the best. As human beings, we are satisfied, but we have the potential to be better.
ORIS: But then again, speaking about the end result of the process, you said that you prefer the house itself to be beyond the artificial. Could you explain this idea?
Sou Fujimoto: It is related to simplicity and complexity. Before the 20th century, artificial things somehow had to have simplicity because the range of information people could control was small, so we had to use the straight line or the 90-degree angle to control situations. Now we have new technology, a new spirit, new everything. People can control complex situations. For me, the most complex things are the natural things, beyond artificial things, so I try to get close to the more natural complexity than the usual artificial simplicity. We should have some kind of simplicity and complexity together, not just a chaotic complexity, then we could have something beyond the usual artificial things.
ORIS: When we speak about the relationship between artificiality and nature, sometimes you use nature as a metaphor for your work, for example, the metaphor of the tree.
Sou Fujimoto: One of the reasons for it was that I was born in northern Japan, the Hokkaido area which is a countryside area, so in my childhood I used to play a lot in the forest. In the forest you are very protected by trees, but at the same time you are not closed in, you are in the open so you can go wherever you like. It is both open and protected, it is one of the fundamental forms of architecture. At the same time, I don’t want to translate natural things straight to architecture. I don’t want to use the shape of the tree itself, or the system of natural situations. Instead, I try to somehow mistranslate natural things to architecture, so a tree is not a tree but I try to take each space on each branch and to create new, more networked relationships.
ORIS: Many of your more recent buildings are literary three-dimensional networks constituted of autonomous elements freely arranged in space. By introducing three-dimensionality, you created denser spatial dispositions. Does it have anything to do with the Japanese situation where land is so valuable and precious you have to develop buildings vertically?
Sou Fujimoto: Yes, exactly. The starting point is really for such practical reasons. One house is under construction with many small floors, stepping three-dimensionally. That house is one of the first houses I designed in Tokyo. The size of the plot is very small, so we thought ‘small site, small house’ is too usual, it’s just a small house, so we decided to create many smaller floors related to each other, like furniture. Then I found we can create such a big size of the three-dimensional space with many steps, so you can feel not one small floor, but a three-dimensional, very big space. We have many different levels, different floor sizes, then you can have your own various areas where you can choose to be. In such a small site, for me, the best strategy is to maximize the smallness, enjoy the smallness, not make the maximal small floor, but to add different dimensions. I don’t want to limit and I try to create more common strategies for architecture itself. Many small floors could be a nice strategy even for a big house because it is based on basic comfortable feelings for people, or basic comfortable relationship to the space.
ORIS: The idea of comfort in Japan is quite different than in Europe, for example. In the apartment house in Tokyo, you introduced the concept of the apartment where you have to go outside to get to another room.
Sou Fujimoto: That Tokyo apartment is rather an exceptional case, but in a sense comfort in Tokyo is a little bit strange. Tokyo is really crowded with small paths and small spaces. I found out that it has different kind of comfort. It is not stuck in a small space but the small, cozy space has no end, it is continuous throughout Tokyo and the whole area. In the Tokyo apartment, I tried to recreate this kind of Tokyo comfort in one apartment house. Each house unit box is very small, but the relationship could have a more varying potential, so you can go up by ladder, or you can go out to step out to the upper house shape. It is not just one small space but you can have many possibilities to walk around in the house. I tried to make it like an extreme Tokyo.
ORIS: It’s a recreation of city conditions and its in-between spaces which are also a part of city life. What about the social space in Tokyo, how does it work? How do the streets and in-between spaces perform their public role?
Sou Fujimoto: It was very interesting to me when I moved to Tokyo for the first time. I had a very small room, but when I go out, the street is also small, everything is so small as well. Inside is my comfortable space and outside is also a cozy space, but of course, we have to share it with other people. It is no problem because it has a winding space, it is not straight. A corner is in front of me, and then I walk for fifty metres and come to another corner. Tokyo has various different areas but I’m talking more about the housing areas, complicated areas. In those areas, public things and private things are well mixed together like a gradual situation. The house is a more private area, the street is a rather public area. Such a nice gradation of things is very interesting to me.
ORIS: Many streets in Tokyo have no names and houses are coded in very complex way, like some sort of sub-species. Could it be that it has something to do with this idea of a gradual, non-hierarchical space? It’s just that you have to engage so much to understand how it works, but you will then also understand the genealogy of a certain area.
Sou Fujimoto: Japanese houses are detached, they have a minimum of fifty centimetres of space between them. Sometimes there is a one-metre gap. A one-metre gap is very similar to the minimum width of the street, so the gap between the house and the street is not so clearly divided, sometimes people can use the gap between the houses in their daily life. There is no definite street, but the street is a slightly wider open space. That is why the street has no name, because the street is just an open space. It was not defined before the houses, but after the arrangement of the houses. That kind of system shows how the Japanese old city is created and how the house and cityscape are mixed together.
ORIS: This is what you introduced in House N, for example, because you have a series of gaps between the exterior and interior, where this delineation between public and private is gradual or blurred.
Sou Fujimoto: These kinds of things were most interesting to me. You can go through someone else’s garden or you can go through the pit between the space of the two houses. It’s a private plot but it’s ok to go through. It’s a kind of a very well mixed situation of public and private. Sometimes it’s very exciting to walk through such a mysterious area to find your own space or to find your own way to walk around the city.
ORIS: You argue that this is one of the new agendas for architecture, to question the relationship between public and private and to see how to reinterpret this concept.
Sou Fujimoto: That is a starting point. I was thinking about the Tokyo house. The boundary between the house and the city is not so clear, so I tried to blur the boundaries with the layers. The layer concept comes from more traditional Japanese architecture. We have a gradient of layers from the city to the house. Of course, that’s related to the more fundamental aspects of architecture. Architecture should have a definite boundary, but at the same time I try to blur the boundary.
ORIS: I think that the Musashino University Library in Tokyo is also about the blurring of the definition of the inside and the outside, but here you deployed idea of a spiral. It’s not like a box in a box, but a spiral disposition.
Sou Fujimoto: The spiral has several different reasons. Usually, this systematic layout is more and more required for a library, but I think that this forest-like, walking around and coming to some unexpected books is very important as well. We try to treat these two opposite aspects as equal and like to combine them together. In the end, we found this spiral shape that could have this kind of nice coexistence, because if you are in the centre, you will see all the categories surrounding you. At the same time, the shape of a spiral is a really primitive plan of a labyrinth. We have many openings, so it could be a labyrinth for people who are not searching for books. At the same time I was thinking that the library is so huge, but when that complex library is made of only one bookshelf, it could be a very funny thing or a very nice thing. I liked the idea that just one long bookshelf could create a really complex library.
ORIS: Do you consider the notion of the beauty of a building, or do you think about shapes as a rational part of architecture and the beauty is some sort of consequence of the design process?
Sou Fujimoto: Honestly, I try to create something that is very beautiful, so in the process of judging each process, my decision is partly deciding whether something is beautiful or not, and partly something that is beyond the usual concept of beauty.
ORIS: When does the diagram appear in process of design? Do you use it as an explanation tool during or after the design process, to understand what was done and sum up what was learned?
Sou Fujimoto: I can say that the diagram does not happen from the very beginning because it is very chaotic. Then in the process we can find some kind of diagrams or simple forms we can develop further in a different way. Of course, after the completion of the project, we can find different aspects, so I think in one project we can find that one diagram per project is not the reality. The project may have one diagram, but with a different viewpoint, with many different understandings possible. Somehow, the diagrams come later.
ORIS: It is also a communication tool and storage of knowledge. It’s a distilled idea of an idea or project.
Sou Fujimoto: The diagrams and key phrases or key sentences come from the project, during the process or afterwards.
ORIS: For example, what was the key sentence for the library?
Sou Fujimoto: The library contains the diagrams of co-existence of searchability and storeability, systematic things and chaotic things. That is one aspect, and another thing is that after completion this library space is a kind of sequence of expectations. Most of the area is blocked by the walls, and some openings could show the area behind. We will then be surrounded not by architecture but the whole expectation, then we could say the architecture is endless. The architecture is stopped and fixed, but your experience is endless. The expectations could create another set of expectations.
ORIS: This is a perfect place for chance encounters, you may accidentally run into the book you need. In a way, the architecture disappears because the environment is dominated by the books or the experience of browsing.
Sou Fujimoto: Architecture that disappears is the ideal architecture.
ORIS: Your architecture has the ability to disappear yet it still discretely orchestrates events. Unlike architecture that is so present that it claims it opens up new possibilities. Can you explain the ‘guruguru’ concept?
Sou Fujimoto: It means ‘spiral’ in Japanese. The first spiral I used was not in the library but a private Spiral House project. In that case I intended to create some kind of continuity or gradient, changing from outside to inside, or from public to private. In the spiral, that house is facing the street, so this area is more open. Gradually it gets deeper inside. The private is not divided by walls but the lengths of the space divided the private and public area, so it had different realizations of layers from the city to the house. That is the beginning of the spiral. In the library we found different aspects of the spiral, a nice co-existence of the systematic and the chaotic. The guruguru spiral concept is growing. It’s one of the most primitive shapes people draw, cave paintings and drawings contained such things. Somehow I like such a really primitive thing to retranslate. Of course, Le Corbusier used spiral shapes for the growing museums, so that is a more systematic translation of the more primitive shape, and here it is a different translation. It’s not a new shape, it is very old, but with a new interpretation, a new innovation of meaning. Guruguru has no meaning, it is an expression of feeling.
ORIS: If you think about the present condition of architecture, which is in a sense deadlocked, I think your work opens up one possible way out of the crisis, because it’s a kind of humble understanding of complexity. How does this coincide with the situation in Japan after the disastrous earthquake? It is going to be a big challenge for architects to recreate what was created more spontaneously. Your research of the translation of Japanese city structures to architecture may be a useful experience for the sensitive reconstruction of devastated areas.
Sou Fujimoto: It’s not only a problem of architecture, but of infrastructure, cityscape, urban planning, ecology. It’s a big challenge. At the same time other Japanese architects and I are powerless when it comes to urban situations, because Tokyo is moving and growing day by day, and architects can do only small plots. We can propose urban visions but they have no power in reality. New developments suddenly happen and sometimes only a single small passage divides the existing wooden house area and the high-rise area. It is a kind of very strangely mixed city. For example, China is now building new cities with a master plan designed by architects. I don’t know whether it is good or bad, but nevertheless, they can do it.
ORIS: Actually, the only thing that is stable in Tokyo are the parks, this is something that doesn’t change in form or position. They are always the same, stable structures. The urban tissue between the parks is constantly changing. We have this peculiar inversion that the solid is changing and the void is stable.
Sou Fujimoto: I am pleased that this kind of mysterious part of Tokyo actually exists. The basic infrastructure, or basic framework of Tokyo is based on the Edo period, the samurai period. It is now a very modern city, but the basic street layout is definitely defined by the Edo period, so we have very strange paths.
ORIS: Can you explain the concept of your award-winning competition entry for the Science Centre in New Belgrade?
Sou Fujimoto: The 20th century science centre was somehow a huge warehouse with many, many scientific things. We thought about what the 21st century science centre could be like, and we thought it could be like a forest, where people can walk around in nature, so the scientific things are not just in a warehouse, but also co-exist with nature. So we made a huge, wall-like patio, 15 metres high and rather thin. It’s a patio, but like a tall tree itself. We arranged the plan of the patio to create various spaces where people can walk around. There is also a huge roof. It is so high that you can’t feel the roof itself. We also put reflecting material so you will see the patios continue to the sky. The library forest is a more intimate forest, and this science centre forest is more like an actual forest. In my mind, there is a kind of contradiction, natural or artificial forest or in-between things. I am currently designing an arts museum in Shanghai, a private museum but very big. It is basically a renovation of an old warehouse. We designed the big entrance space of this arts museum as a forest. We used actual trees, so from the outside it looks like an actual forest. We used very simple glass walls to create the interior, the roof is also made of glass, and the museum again has many patios. The trees are on the patios, but we have a lot of glass so the reflection and the real trees can really emphasize or maximize the forest. A forest beyond a forest.
ORIS: Is it a forest that becomes a cloud?
Sou Fujimoto: A forest-cloud. A cloud is in a sense a kind of ideal architecture, because it has fewer boundaries, but it has an area and it has a gradient, different variations of dense or less dense areas, bright areas or dark areas, so a cloud is like that for me. So how do you create cloud-like things? It’s a nice challenge to inspire new architecture.
ORIS: The concept of the cloud came from the Japanese tradition of graphic representation of space. In a single image you have events which are taking place at different times.
Sou Fujimoto: It’s a very typical but very strange Japanese traditional painting. The whole area is covered by a cloud and through the cloud you can see various scenes. For Japanese people, the cloud is like that. It could provide the framework for your experience. The whole thing is defined by the cloud, but the cloud itself is almost white, nothing. Nothing-like things can create a basic framework for your experience. It is a nice inspiration.
ORIS: You often refer to computer science and IT technology. I think this traditional concept of the cloud really relates well to cloud computing. It’s blurry, it has boundaries and it gives you a framework to experience. Like swarm intelligence.
Sou Fujimoto: I think it’s very interesting because now the cloud means not only a natural cloud, but a kind of networking thing. I like this kind of new thing, Internet things, because when something new happens then architecture could change somehow, ecological things or IT things could change architecture somehow because they could change people’s behaviour. I am very conscious about such kinds of new things, but I don’t react too directly, I like to find fundamental meanings or aspects of such new things. The interesting thing about cloud computing is that we cannot see the cloud computing itself, so it’s a big challenge how to translate such invisible things to visible architecture. We cannot see them but we have to find a way to see them.