Skip to content

That First Idea

Post date:

Architectural ideas aren’t myths like The Architectural Imagination. Architects have architectural ideas in exactly the same way cooks and chefs have food ideas about ingredients, their combinations and processes and document them in things called recipes, and in exactly the same way composers have musical ideas about melodies and instrumentations and document them in things called scores.

This semester’s first year final project was for “a little house” of about 60 sq.m and, as an illustration, something along the lines of Le Corbusier’s 1925 “Little House” on the shore of Lake Geneva. Living room, bedroom, kitchen, bathroom, space for a guest. Sweet. I never knew this was yet another house for his beleagured parents. This time around LC found the site, sketched the plan and then buggered off back to Paris, presumably to work on being famous. It’s a familiar situation, except today, the local contractors left to deal with things like construction would get to be known as architects of record.

Only two years earlier LC had declared “The plan is the generator!” as if it were insight. Patrick Schumacher claimed, au contraire, the random “irritant” is where the design process begins. Who to believe? A plan was a good way to start even before it became something Le Corbusier said. But how to generate a plan? How does a person have an architectural idea? And what is one anyway? More to the point, is it even possible to teach students how to have one? It’s not as if all one has to do is lean over a student at their desk and command them to “Be creative!”. I don’t know. What I do know is that, after graduating, more than one student has said to me the thing they remember most is me telling them “Don’t fall in love with your first idea. It’s unlikely to be your best one.

There were plenty of nearby sites to choose from and most people, me included, chose some effortlessly picturesque one such as these, with the river to the south and the village behind.

I know I shouldn’t be constantly amazed how strongly the reality of Chinese landscape concurs with my image of what I expect it to be. I’m so used to representations letting me down but my experience so far of China is they don’t. Many days I see mountain ridges layered against the mist. I’ve looked up and seen cranes their wings describing a perfect circle. I’ve looked down at a river to see two carp swimming in yin-yang formation. Between my office and studio I see frogs resting on waterlily leaves. All this would happen without me. I used to think Chinese paintings were abstractions but they’re really just painterly recordings. Post-modern me needed reminding that reality is independent of and more important than representations of it, not vice-versa.

Such reveries are all well and good but that building isn’t going to design itself! HOW TO START? And how to teach students how to start? I learned it’s not that easy following one’s own advice but how does one even have a first idea and then not fall in love with it. Like first loves, it’s better to not get too attached so perhaps starting with lowered expectations is better. Uncertain, and feeling a pressure to do something, I began by drawing the first thing that came into my head, just to get the brain and hand used to working together. I said, Why not try miniaturizing some famous house and see where that goes? In this next scan you can see Richard Meier’s Douglas House reworked as a little house – ugh – except it didn’t really work`

“Think about what you’d like to see there” I suggested next. “Imagine someplace you’d like to be, something you’d enjoy seeing there.” To illustrate, I channeled Mario Botta into a miniature castle half in the water, half not. It kind of worked but the external staircase was cheating given the 60 sq.m area limit. Nevertheless, I imagined myself moving between the morning terrace and the evening terrace, conveniently forgetting how cold it gets here between November and April. I was succumbing to that architectural trope, The Summer Weekend House.

A student told me it was bad feng shui to have the downstairs bathroom door so close to the entrance. I’d also forgotten to include the space for the guest. Oops.I’m not making myself look very good here.

Another idea you can see on that first page was a Philip Johnson / Tadao Ando mashup. Ando and Johnson might seem unlikely bedfellows but Ando’s Sumiyoshi and Johnson’s Rockerfella Guest aren’t exactly apples and oranges. I imagined myself sitting on the terrace with a Campari and orange at sunset watching the river flow by.

The rest of the time I’d be upstairs at my desk channeling Curzio Malaparte at his magnificent window in his magnificent study of his magnificent house.

Also on that first page that first day I drew a variation with a curved metal roof and a low Japanese tea-house window like Shinohara’s House in Kobe, as well as a roof and stair in the style of Alberto Campo-Baeza. All I was saying was draw stuff, see what happens, and what it makes you think of next. I was still loving my external stair and imagining my imaginary self lucky to not be living in Ando’s Sumiyoshi House in some airless Osaka summer.

The next weekend we had a field trip to look at some Chinese covered corridor bridges like this one, all constructed without nails, screws, welds or metal of any kind. They’re wondrous things, confronting physical forces with the fluidity and grace of dragons.

Inspired in class Monday, I squared up my mini castle, made the stair internal, and tried out walls a combination of Chinese dry rock walling, brick and vertical timber below a tiled roof. But it wasn’t right, probably because it looked like a farmhouse when it was anything but. Plus, the staircase sucked. I’d managed to fit an internal staircase into the layout and using as little additional space as possible but it didn’t seem like a very good use of six out of the permitted 60 square metres.

Meanwhile, my mashup wasn’t going away. Misfit me was now liking the idea of three out of four rooms with usage undefined.

The only thing to decide was which space to make the bathroom/kitchen but, before I could decide, I became disenchanted with the whole external stair thing. This idea had gone as far as it could. It had run its course. The next Monday I told the class I’d started again, and with how should have perhaps started in the first place – with the plan. I showed them a plan and sketch I’d drawn on an A5 postcard and explained that when I wake up, I usually go to the bathroom and wash my face and, on the way to the kitchen, check the computer, and then make myself a cup of tea and either drink it on the balcony or at my desk. At night, something like the reverse happens and it generates a plan like this, now with added vegetables.

I said I wanted to see the morning light illuminate the bedroom ceiling and for the setting sun to bounce off the ceiling of the living room. It was an okay design and with a roof owing something to an earlier project of mine. I admit, it was a bit dull. Pressure.

I thought about roof shapes and materials, eventually settling on a curved metal roof shamelessly borrowing from Shinohara’s House in Yokohama even though I’d never particularly liked or understood it. Anyway, unashamed appropriation is always acceptable when a former student does it, perhaps because it perpetrates the myth that architectural talent can be transmitted by proximity alone.

The blocky massing wasn’t doing it for me. It might have been great when I was a first year student but I now needed to present something more ambitious. Still, the best I could do was some Shinohara reimagining with added efficiency. I can’t say I haven’t been influenced by all this but when does aesthetic reductionism start being art and stop being art? It probably has to do with the number and artistic aspiration/pretentiousness of the ideas either evoked or meant to be evoked.

I placed the kitchen in front of the bathroom and put the guest area above. The little house became littler. There was still some pondering of windows to be done. I think a little house shouldn’t be burdened with big ideas and this is why, to my mind, Shinohara’s House with an Earthen Floor will always be superior to Umbrella House or Prism House or House in Yokohama, even if we were never told what those big ideas were. The pizza-slice window is my own invention. I make no apologies. Were he to have seen it, Shinohara would’ve maybe raised an eyebrow and left it to us students to infer what it meant.

The plan looked like coming in bang on 60m2 with about 8% circulation and the only dedicated area being the small space between the entrance and the stair. The plan may still be the generator but that all depends upon what generates the plan. In this case, the plan was generated by sequencing the rituals of living. Just when it was time to fix and dimension the geometry and find out how accurate my sketches were, I received the following email.

Hi Graham,

People in grad school (I’m in the first year of a masters program for architecture, not having having a bachelor’s of architecture) and people on the internet seem to believe that Revit or ArchiCad is the future. I get that it’s likely a necessary skill set for getting a job, but it’s virtues beyond possibly making the construction of large projects like hospitals, hotels, airports, etc seem overblown to me.
I’m stupid but it seems like BIM can’t possibly ensure the bright future everyone seems to pretend it can. …It seems like just another tool it’s possible to be stupid or smart with.
I would enjoy reading any thoughts you have about BIM…is it an architectural myth?…
Many thanks for your blog which I consider an invaluable supplement to my graduate education.


Let alone the future, the present state of architectural production is not offices staffed with multi-skilled architects but offices with greater automation, a strong division of labour brought on by those various automations and, following on from that, lower overheads (ditto). BIM fits this bill precisely. Over the last few millennia, many fine buildings have been built without BIM but not with the short design and construction periods that corporate profits demand. Like automation in general, CAD and BIM enable building outcomes and products that would otherwise not have been possible. In 2003 (AR, April) I wrote the following in response to Charles Jencks gushing over the use of computers on Gehry’s Bilbao Guggenheim and cheerleading it as some new paradigm for architecture.

I still believe this, and I agree that BIM as just another tool to be smart or stupid with. Revit will alert you if your A/C duct clashes with a column or beam but it’ll also let you design as if gravity didn’t exist. ArchiCAD was the first CAD program with integrated BIM and also the first program with integrated 3D. I’ve been using it since about 1995 to confirm ideas imagined in my head and sketched by hand. These are my first tries.

The history of the project is a story in itself but a friend owned the rights to build on the roof of an unlovely block of apartments in west London and was wanting a proposal with which to approach planners. The first proposal was drawn in tracing paper on A3 enlargements of photographs. The second proposal with replaced brickwork was pencil and colored pencil on cartridge paper. We called it cutting and pasting.

Here’s the same thing modeled as proof of concept – at least visually – but even so, it’s no better as communication. When showing first proposals to planners, I’ve always had better results presenting hand-drawn, undercooked proposals because the discussion goes more smoothly when you acknowledge the agency the municipality has. I’m pleased I can show you pre- and post- images of these two proposals but, even so, they’re both just alternate representations of a building I once hoped would exist. They were only ever meant to represent the thing for a little while.

Having said that, “drawing it up” and settling geometries and dimensions is never just that. The thinking and designing continues but with shifted emphases and more attention to detail. At some point the designing has to end and the documentation begin but designing never ends as long there’s an opportunity to improve it. Here’s the current final iteration of The Little House. Any changes didn’t come from the design itself but from trying to settle it. I’m pleased with how it turned out. These two drawings are either the end of the process or the beginning of something else.

  • The kitchen was too deep so I shrank it and fitted the stair between two walls, much as Shinohara would have done. [There are only a few Shinohara staircases not sandwiched between two walls.]
  • The stair is steep and has no upper landing. You step off the stair and directly up into the room. It’s totally non-compliant Chinese rural vernacular.
  • I simplified the rear elevation, added some more windows.
  • I kept the Shinohara pizza windows on the west elevation knowing full well a builder will cost them out of existence. I’m not wedded to them – they’re sacrificial.
  • Guttering and downpipes call attention to themselves but Wenzhou where I now live receives some serious rain. I’m not playing.
  • The flat roof needs some drainage. I need some advice on roof construction.
  • I left the ceiling heights as they were. The kitchen and bathroom are minimal, the bedroom a cabin, the upstairs a Japanese tea ceremony room.

Thanks to everyone, misfits’ architecture is eleven today.



  • This is the quintessential primer. I learned more in this 10 minute read, then what I recall learning in 10 weeks of arc 101. Your final product should be in a tiny-living plans catalog. I like the way you clarified how you get into the loft spaces. Well done mate.

  • Hi Mr Mc Kay,

    thank you for keeping at these blog posts. i have been a long time lurker and avid reader. But this essay probably hit closer to the feels for me. Having practiced in a small time boutique practice at one point. And having a very similar first year assignment.

    Having the first idea depends on so many factors and it is probably difficult to grade the students work other than the oft quoted “Overall Quality”. This confuses the student but this is the part that makes architecture education challenging.. by helping the students quickly realize that developing the first idea is a really a skill of reading the air and make inferences to propose for the tastes of the examiner, clients or even the student themselves.

    The process is often like the process you have illustrated above. Except, in a real life scenario, there would be many back and forth dialogue between the clients and the architects and having a feel for the clients prepositions.

    In terms of a student assignment, perhaps there could be a fleshing out on the proposed experiential events like the event with the Campari and orange essay for instance, These proposals can be potentially extremely varied.. but also give meaning to the various idiosyncrasies in the proposed design.