Interview: Sou Fujimoto 04.06.13

Icon119 Fujimoto-inside1

Serpentine Pavilion exterior, 2013 (image: Sou Fujimoto Architects)

The architect of the 2013 Serpentine Pavilion talked to Icon about the qualities of clouds, the integration of landscapes and buildings and his quest for serendipity.

ICON Your work often refers to natural imagery – clouds, mountains, forests and so on. How has this informed your design for the Serpentine Pavilion?

SF This time the pavilion is like a cloud. I was born in the Hokkaido area, which is in the countryside, and then I moved to Tokyo. And I found a surprising similarity between a natural forest and the Tokyo situation. The forest has a nice complexity and diversity and Tokyo has similar complexity, but made by artificial things – no clear order, winding streets – it's like a forest to me. I found it interesting to think about architecture in such a way.

So forests were the very origin of my architectural thinking. And the forest and the cloud are similar – the cloud is soft, like air, but has a diversity of densities. It is like architecture, I think – when it's most dense, it's like a wall, like architecture, and if it is less dense then it is more like space, but there are no clear boundaries between the space and the structure; it is more melting together as an integrated whole.

At the pavilion, people come into a cloud-like space, with a nice distribution of different densities of structure. You can behave as you like, inspired by the bright areas, the dark areas, the landscape-like hilly space, or cosy areas, or openness – there is a lot of different space. The cloud has a soft order, almost no order, and it is quite similar to the forest and Tokyo contrast.

ICON You've spoken in the past of making landscapes rather than buildings. Why does that appeal to you? Architects might be expected to prefer to think in terms of buildings.

SF I like to think about architecture as a wider integration of things. I don't want to divide landscape and architecture or cityscape and architecture [but instead] melt them together to provide a variety of fields for people to behave as they like. In the Tokyo Apartments (2010, Icon 085), the main topic was stacking the house shapes, but the important point was that we put the staircases winding up to the houses.

This was like an extension of a small Tokyo pathway going up into houses. The Tokyo landscape was integrated into the architecture. People's experience and behaviour are not divided by buildings and furniture and landscape; it's more continuous, with them all blurring and ambiguous. So I like to create a field for people to feel comfortable ... One method we use is to divide the spaces into smaller pieces, then we can get different scales and have nice in-between experiences between landscape and architecture.

Icon119 Fujimoto-inside2

The Tokyo Apartments, 2010 (Daici Ano/FWD)

ICON This is your first building in Britain, your first building in London – is there anything you've drawn from London as a location?

SF Of course this location is very special – at the Serpentine there's a really strong context and history. [There is a] beautiful contrast between the surrounding nature and the pavilions. [The grid that underlies the pavilion design] is so different as an order from the surrounding nature, but if it is made in a delicate way, it can be similar to the complexity of trees. It has a continuity with the surroundings.

ICON In England there is a tradition of creating wholly artificial landscapes that are meant to appear "natural" – and that seems to me quite close to some aspects of Japanese culture.

SF There are similarities and differences. I was impressed by Hyde Park and Kensington Gardens – sometimes the geometry is quite clear, but the experience is made by the trees, and their shape is quite complex and natural. So there's a nice, clear coexistence of natural orders and artificial phenomena together.

In Japanese gardens it's more subtle – it looks not too designed, but it is designed. My attitude is more from Japanese culture, creating order that vanishes into natural things. The clear grids, so strong and white, I hope, are sometimes so delicate they vanish into their surroundings [and] the greens behind appear – a nice coexistence, rather more like British landscape culture.

Icon119 Fujimoto-inside3

House N, Tokyo, 2010 (image: Iwan Baan)

ICON Where do you see yourself in relation to the metabolist movement in Japan of the 1960s? They also had an interest in natural forms and building up complex structures from simple units.

SF In my school days metabolism was almost forgotten, actually. Afterwards I learned about them. But for me the metabolist method is too strong. They have really strong cores, like a tree, and the branches are made of the repetitive units – so it might be inspired by a tree but it's more like a machine. I was more influenced by weak networks of independent small pieces coming together, like the Tokyo situation itself, or a traditional Japanese garden. For the pavilion, there is no clear hierarchy, the grid is more flat and transparent. It is strong but it is very weak.

ICON When you try to draw out the sense of the natural from the unnatural landscape of the city, this spirit of disorder, it seems to me that you're reaching for a fundamental but hard-to-define element within the urban-ness.

SF It's a fundamental question. I'm still wondering what it is. Tokyo is not chaos but it's not ordered. I try to recreate such a mixture in architectural situations. For me that is breaking the various boundaries between order and disorder, between nature and the artificial. [We must] redefine order/disorder, nature/architecture, outside/inside, city/landscape, architecture/furniture – we have many, many definitions, but if we can reinterpret these divided pieces together to go beyond such invented definitions, then I believe we can create something new, and something fundamental.

Icon119 Fujimoto-inside4 copy

House NA in Tokyo, 2011, based on the idea of living in a tree (image: Iwan Baan)

ICON That must be very difficult – almost paradoxical – for an architect to consciously strive to create the unconscious form of disorder.

SF We've been struggling with it for years. For the Children's Centre for Psychiatric Rehabilitation (Hokkaido, 2008), the layout looks like disorder, but each cube has the same size, and the "random" layout is actually intended. In the case of the pavilion, the whole shape is an ambiguous shape, so how do we control such an ambiguous shape? It is like a bonsai tree, sometimes you cut it, but you cannot completely control it. Trees grow as they like. You cut according to how they grow. They are half natural, half artificial. And in architectural design it is wholly artificial, but [we try to create] a natural historical process, [like an] emerging village. Gradually, day by day, the design changes.

ICON What did that mean on a practical level? How was the form developed in the studio? Was it with the hands, sculpturally, or more mathematical, systematic?

SF Not so systematic. More historical. From a simple bowl shape, we added an entrance, we twisted some things, altered it to fit requirements, and then maybe in an accident somebody twists this shape [laughs]. Every small change has a reason, but then as a whole I could not see any reason to make it like this.

ICON Almost like an effort to develop it unconsciously, without ...

SF... without any final intentions, yes. Many small intentions, and practical requirements, and sometimes just personal inspirations. Tokyo is fascinating mostly because it has a history – each layer had intentions, had plans, but some of the plans are interrupted by other things, leading to a nice mixture. It's impossible for architectural design to imitate such a history in an artificial way, but I tried to open the process within my office, setting no clear goal for the forms ... and then see how it likes to grow.

 

Words

Will Wiles

quotes story

From a simple bowl shape, we added an entrance, we twisted some things, altered it to fit requirements

Leave a comment

Click to show