Repeating Crevice is the English title Kazuo Shinohara gave to a 1970 house that, in Japanese, is known as 同相の谷, dōsō no tani (“In-phase Valleys”). The drafting style of the plans below shows they came from one of the two early books that led me astray.

• 

• 

For many years I thought of Repeating Crevice the way it was presented – as an architectural exploration into domestic space as Art. If Shinohara was aware of having designed into it certain possibilities beyond that, he never let on. The approx. 12 m x 12 m footprint made me recall The Expansible Home and want to revisit Repeating Crevice and see if it has any lessons for us today. Before I do, you might need to work out what’s going on with these plans.

This might help.

So might these.

This definitely will.

• The (green) entrance hallway is shared space.
• The (lilac) larger apartment on the ground floor has living areas on the left and, on the right, a bedroom, bathroom and some unidentifiable space accessed via the (blue) downstairs space that is semi-private because it’s overlooked from the (red) upstairs semi-private space.

• 

• 

• The smaller (pink) apartment is also split in two with its living and sleeping areas separated by the upstairs (red) semi-private space.
• A window in the small lobby when entering the (pink) living areas is open to the double-height space of the (lilac) living room.
• The six-mat Japanese-style room is a shared space accessed by the downstairs (blue) semi-private space or directly from the upstairs apartment.

• • •

Here’s what I mean about The Expansible Home.

• 

• 

The arrangement on the left is an apartment suited to, say, a small family. It’s no inconvenience to pass through the living room in order to access the kitchen. In both apartments, the occupants of each bedroom have equal access to dining, living and cooking areas. It’s not just about the spaces though. The arrangement on the right is more suited to a houseshare or co-housing because of the different ways the occupants move about it. The plan is not generated around the usual “promenade” from entry to living room. Occupants can enter and leave the apartment without having to pass through living areas.  They can also move around the apartment in response to the presence or absence of others in those areas.

There are limits to how far the hotel model can be applied to co-housing.
Architects, developers, and probably even the co-housed have come to believe a successful development involves a groovily-decorated shared space to which people will gravitate and do whatever it is they’re supposed to do there.

It’s likely that such spaces have come to represent co-living while not doing very much to enhance it beyond representing the minimum expected level of amentity. The current architectural manifestation of co-living has quickly settled on articulating the binary states of together/alone and connecting them by a corridor that represents dead time as it’s time spent in neither of the only two states imaginable. This is the hotel model.

True, one could meet someone and have a conversation in those corridors and, convivial though it may be, it’d be something to pass that time. Certain 1920s Soviet communal houses had heated corridors and seating to encourage the use of shared circulation space as shared amenity space. Such an arrangement means that obligations to society (or at least to be social) exist the moment you leave your apartment. That future never happened, but co-living using the hotel model is now with us in a big way and people are expected to be either together or alone. The absence of a buffer zone separating the two states means neither can be anticipated. Co-housing along the lines of the hotel model could quickly become tedious, onerous.

The hotel model is appropriate for short-term occupation by strangers with things to do elsewhere or for people joined in common purpose or mentality. The hotel model has never been tested for long-term co-housing as a substitute for housing types no longer accessible.

The socially useful and necessary idea of co-housing has already begun to be negatively regarded but the flaws being pointed out are not with the idea but with the model chosen to implement it. The only lesson of lasting value that streets-in-the-sky,  Pruitt Igoe, tower block council housing and Brutalism in general taught us is that any socially useful idea will be deemed a failure once the flawed models chosen to implement them have been exploited to the max. Co-housing is currently being set up to fail.

Sometimes just knowing someone else is at home is sufficient.

Repeating Crevice was designed to be occupied by two generations of the same family and, as such, is a form of co-housing. My hunch is that it contains ideas for how any group of people might live together with the advantages/comfort of doing so. It seems to allow for flexible degrees of awareness of being together or being alone.

Both apartments are split by semi-private spaces that are necessary circulation spaces.
The upper apartment is most likely the apartment for the parent/s but I only say that because a window from the bedroom overlooks the entrance hall. Because the upper apartment’s semi-private space (that is also well travelled) overlooks that of the lower, the person in the upper apartment is more likely to see (and thus derive comfort from) occasional activity in the space below. A person sitting in that chair in the header photograph would be aware of all people arriving and leaving the house. When all are at home, they would also know where in the house they were for there are no alternate routes, but that’s all they get to know. Some privacy lost means other privacy kept. Going from one part of a house to another is generally not a fully private activity but doesn’t have to be a fully social one either.

This runs counter to today’s thinking that holds any and all opportunity for interaction to be A Good Thing and the more of it the better. It’s bad enough that corporations find potential to monetize forced interaction in corporate environments but forcing people to be social in domestic ones could just be a new kind of hell no less inhumane than alienating them. Together/alone is another false binary. We’re encouraged to think of solitude as anti-social.

Repeating Crevice, the house, is a largely internal environment.
It has windows to the front and a some garden to one side but all the architectural action is internal. It doesn’t require an external view to give its internal spaces meaning. This is a useful trait for living spaces to have as buildings tend to not have 360° unobstructed views of something nice. If you refer back to the plans above, you’ll see the upstairs bedroom window is (exactly) 1m away from an upstairs window of its neighbour.

Entry is shared and bathrooms and kitchens are stacked. Good. This all means that

What we have is a potential model for vertical co-housing that is not based on the hotel/hostel typology.  

Assumptions

1. The semi-private spaces are also semi-public spaces in that somebody else’s visitors may appear.
2. The house is lived in as a houseshare. The people know each other well and their interests are shared, they look out for each other, and trust each other. The front door is the only lockable one. If this weren’t the case, we would have a house in multiple occupation, or something pretending not to be a hotel/hostel.
3. The two-storey maisonette has a fire-escape stair and a fire-fighting lobby so it can be within a multi-storey building.
   

   

   • • •

Repeated Crevice, Revisited

• The fire-escape stair and the firefighting lobby means the dwelling can be vertically repeated as far as the structure and two elevators permit.
• The maisonette can house up to eight people.
• The elevator and fire stair are not accessible from the upper level. Instead, the spaces is used as a spare room for possible use as a guest room or office.
• Upstairs and downstairs have different doors leading to/from the (fire-protected) elevator lobby.
• Downstairs rooms can directly access the kitchen/dining room but upper rooms can directly access the living room.
• Instead of Together/Alone there is now Together (green), Alone (lilac), Semi-alone (pink) areas and Semi-together (red).

• The kitchen/dining, laundry and guest toilet are shared by all occupants, as of course is the elevator lobby.
• People are not forced to be social. Everyone can enter and leave without having to pass through the kitchen/dining or the living room as the elevator lobby leads to the hallway linking all these spaces, but also links separately to the upstairs hallway. A person whose room is on the ground floor could go directly to their room via the hallway, or bypass it by going to the living room via the upstairs hall. A person whose room is on the upper floor can access the kitchen/dining room without having to pass through the living room.
• Staircases are used selectively, but not exclusively so. The upper-level people are more likely to use the stairs by the elevator to get to their room but are more likely to use the stairs from the living room to access the kitchen. The lower-level people do not need to use any stairs to access their rooms but are more likely to use the stairs by the kitchen to access the living room.
• Internal windows enable persons in one part of the dwelling to have an awareness of what else is happening. It is important those openings be glazed. Windows in the semi-private (pink) corridors alert persons moving in either direction to potential social situations. Without leaving a semi-private zone, it is possible to know (at night) if the living room is occupied. No such awareness is possible for the alone zones.
• From either side, the windows onto the two-storey hallway enable everyone to have an awareness of living with other people.

How they choose to act on that awareness is up to them.

[cite]