Skip to content

The Houses of Arata Isozaki

Post date:

The Nakayama House is Isozaki’s first recorded house, completed in 1964 one year after he left Tange’s office to start his own. It was later demolished and a facsimile built in 1998 at Akiyoshidai International Art Village.

By 1964 it was already expected of Japanese architects to have elliptical theories about their work and how it should be interpreted. The idea being explored in this house is said to be a layout unrestrained by the geometry of the building. It’s clearer from these next images but why the house needs to be raised to do this I don’t know. It could be because in Japan raised buildings indicate contents worth protecting. I suspect it’s concrete to differentiate Isozaki from the Japanese tradition of timber buildings and thus differentiate him from Tange’s Own House (normally attributed to Tange) 1953 and Shinohara’s 1954 House in Kugayama, both of which did the Japanese raised thing on wood and metal columns respectively. For 1964, the concrete and the severe geometry seem prescient of 1970s protective/defensive enclosed houses such as Ando’s 1976 Azuma House and Shinohara’s House in Higashitamagawa.

But back to the stated theme, it’s easy enough to contrive the internal layout of a building to be orderly but a designed disorderliness is just as contrived as a forced fit. The exhibition notes say this design was intended as a criticism of architecture’s reliance on flat two-dimensional plans. Perhaps, but it’d be nice to be able to find some interior photographs so we could check how this disjunction between plan and building played out. It’s an interesting idea but say, unlike Shinohara’s 1974 House in Uehara, I can’t see any compelling real, imagined or assumed reason why this shape has to be the way it is. An infinite number of other enclosures could have illustrated the same idea. Maybe it’s a criticism of contrived difference? In passing and in retrospect, architecture’s reliance on the plan as the representation of a user is not addressed, so what we’re left with is the disjunction between a representation of architecture and a representation of a user. Sixty years on, I don’t suppose it matters, but the idea of a user structuring their own lives within an enclosure remains a potentially useful idea even if it’s not what’s being represented here.

Architect’s first houses are never far from home so it’s not surprising that Nakayama House is in Isozaki’s hometown of Oita City, a city that did Isozaki well with the 1959-1960 Oita Medical Hall and Takwasakiyama Manju Temple Plan and the 1962-1966 Oita Prefectural Library all begun while he was working for Tange. Iwata GIrtls’ High School all either completed or begun while Isozaki was working for Tange. There’s also the 1974 Oita Fujimi Country Clubhouse, as well as the Toyonokuni Information Library and Yutaka’s National Information Library, both completed in 1995. The city of Oita is something of an Isozaki pilgrimage site.

There’s one pther Isozaki house in Oita but I’ll keep this chronological according to the year of design in the hope we can see some development of ideas. There is a sequence for all but the last two, with the design of Bjornson Studio begun one year before that of Nakagami House, but construction completed three years after.

–1964Nakayama HouseOita, Japan
1973–1975Yano HouseKanagawa, Japan
1976–1977Hayashi HouseFukuyama, Japan
1976–1977Kajima HouseTokyo, Japan
1977–1978Karashima HouseOita, Tokyo
1977–1979Aoki HouseTokyo, Japan
1979–1980Irahara HouseTokyo, Japan
1981–1986Bjornson StudioCalifornia, USA
1982–1983Nakagami HouseFukui, Japan

Vaulted roofs were a feature not only of Isozaki’s houses designed from 1973 onwards but also with the Oita Fujimi Country Clubhouse above, from 1974. Most of the houses also have a double-height space from which one can appreciate the vaulted or sometimes groin vaulted roof. A vaulted roof is a nice way to finish a building but vaulted roofs aren’t a part of the Japanese tradition in architecture. I suspect Isozaki was positioning himself to a Japanese audience, as not being a traditionalist. Vaults aren’t what anyone thinks of as Japanese architecture.

Yano House 1973–1975

These plans, elevations and sections let us have a better look at what’s happening. Entry on the side, left to the dining area half under that floor shown by the dashed line and up to what is probably some sleeping areas opening onto the rooftop terrace. The axonometric shows the stairs have some tricky winders because they’re not beneath the vault when they begin to rise. The photograph on the right is taken from the stair landing looking across the front of the house. What we know so far is that Isozaki likes vaults and groin vaults on symmetrical plans with severe geometries. A single semicircular wall is good in plan and a double-height space above or partially above the living room is good in section..

Hayashi House 1976–1977

The Hayashi House is Isozaki’s largest and its roof is made up of four vaults forming the shape of a squared-off capital letter “Y”. Here, the fork above the living room is the double height vaulted space. The processional entry seems a bit dated but traditionally, the reception room Japanese houses is often accessed by a circuitous route to make the guest feel as if they are being daken deep into the house. This contrasts with the Middle eastern or even the Western reception room at the very front of the3 house next to the entrance. Midway along that ramped route is a stair to the upper floor and leads to a bridge beneath the curved barrel vault. It’s a large house but without much living space.

Kaijima House 1976–1977

Karashima House 1977–1978

This is Isozai’s other house in Oita. Here’s three images of it from Daniel Munteau’s excellent tumblr blog OfHouses. Just when I decided write about the houses of Arata Isozaki, I discover that Daniel had already featured seven of them back in 2017!

The blogger responsible for Tate-Mono [building, in Japanese] was fortunate enough to be shown through the house by its owner. Otherwise, these photos above would be the only ones we have. [] They take a good photo. The house is either also n office, a clinic or has a lot of visitors because it’s unusual in a house to have a WC sign on the toilet.

Aoki House 1977–1978

The 1977 Aoki House you can see on the left middle of this next model. It’s the one that I mentioned in a previous post as about to be sold and most likely demolished. The listing mentioned a vaulted roof you can just make out in the photo.

Here’s three images of it taken between 1977 and 2017, a forty year period in which the building on the left changed three times and the one on the right twice.

you can find more images here. Many of these photographs are from OfHouses and many of the plans, sections and elevations are from two sites that are keeping the memory of these houses alive even though nobody seems to went them to. The OfHouses blog remembers old, forgotten houses but a site that deals with old, forgotten projects by major architects. This is Izozaki’s Tsukuba Center Building from the period 1979–1983, the same time I was at Shinohara’s. Everybody was talking about it at the time. It was big.

Irahara House 1979–1980

House of Nine Squares 1980

This is an odd one. It doesn’t appear in the Shanghai exhibition or the company website but there is description of it on the Art Forum website, some plans and elevations and a section of it on the Archweb site, and a 90° axonometric of it on the Archive of Affinities Tumblr blog, but no evidence of it having been built. The description and the rationale for it on the ArtForum article is revealing though, and finally we get to the point of all this vaulting and symmetry. It appears the design of the house was commissioned for an exhibition titled “Houses for Sale”, of buildable “avant-garde” architecture that promised potential purchasers yet unbuilt masterpieces. Every now and then something like this comes along. The last one I know of was Ordos.

Bjornson Studio 1981–1986

Nakagami House 1982-1983

It’s just as difficult to place these houses in the history of modern Japanese architecture as it is in Isozaki’s own career trajectory. These nine houses were included in the recent Shanghai exhibition, although compressed into a single model tucked away in the corner indicated as E1 in the exhibition plan as E1.

If it’s difficult to give these houses a prominent place in Isozaki’s own career trajectory, then it’s impossible to find them given any importance in the company profile.

The Californian house, Bjornson Studio, is the only one of the nine houses shown on the Arata Isozaki and Associates’ website. I’m not saying it was a freebie, but its acknowledgement probably has something to do with its design and construction period being identical to that of the Museum of Contemporary Art, Los Angeles. Here’s two views in which we can see back to the 1964 Nakayama House and forward to the 1991 Team Disney. In this period of business growth, Arata Isozaki and Associates were the mildly edgy but nevertheless safe pair of hands that boards of directors warm to.

As I mentioned in The Floating World, Part II, Isozaki’s 1986 S-House is not listed on the Arata Isozaki and Associates’ website either and, given the other projects on th go at the time, it’s easy to imagine nobody in the office had much time for this house. Before going back to that Japan Property Center article, let’s take a look at that list of company projects for around 1986. Obviously a case of bigger fish to fry.

Hmm. “Isozaki has designed two buildings in Katsuyama – this house and the nearby Nakagami Residence Hall (1983).” The “Nakagami House” we’ve already seen. I’m intrigued by the report saying the owner of S-House approaching city officials to sell the land as it was adjacent to the Katsuyama City Hall. The expected value of 6-8 million yen is not that much, but if the house’s total floor area is only 142 sq.m then it can’t be that much and ’s not much land – maybe 10m x 10m in which ten cars could be parked in two end-to-end rows of four.

So what are we to take away from all this? Should we even care about these buildings if even their architect doesn’t acknowledge them? Just looking at that list above, it’s not as if Isozaki’s office was stuck for work throughout the 1970s. Given the circumstances around the construction of Bjornson Studio and possibly S-House, all these houses just may have been throwaway projects to catch some bigger fish. My reasoning is simple. Bigger fish were definitely caught, and these minor projects were definitely thrown away once they’d served their purpose. To put it crudely, they were bait and vehicles for fame only incidentally. The fact that plans and interior images are so few and hard to come by, suggests we already knew this. Nobody is poring over plans and photographs of these houses that did their job whatever it was and were then allowed to be quietly forgotten.

• • •