Image Image Image Image Image Image Image Image Image Image

Bartlett School of Architecture, UCL

Scroll to top


No Comments

Bend Don’t Break: A lesson I’ve learnt at the IAL

I am, by profession, an architect. I was trained to think that the world has a problem, that it is broken, and somehow we can fix it. Wooden beams sag and metal sheets warp. Time shapes materials in ways which we, as citizens of Utopia, do not embrace. Unlike our three-dimensional models, reality is messy and inconsistent. What worried me most is that what I was designing and what I was building, behaved completely different.

Since I started my current research, I started to realise that the world is not broken, it is bent. The wooden beams were not broken, they were telling a story by adapting to their realities. Bend don’t break. Suddenly everything fit into place. I shall focus less on the fictional breaking, and more on embracing and guiding the bending. Embrace the error, embrace the inconsistency. However, as an architect, acknowledging the inconsistency was not enough, I wanted to measure it and quantify it. If reality is messy, how come my calculations are so clean and precise? The problem does not lie in the error, but in the perfect ideal. So I started to fiddle with sensors and feedback loops, in order to transfer this messiness into the digital domain.

I was once told that Calatrava produces additional sketches after his details are built. In a linear design methodology this is absolute nonsense. Sketch, drawing, detail, building. This is what I was used to. However, the feedback loop questions the very principles of a linear methodology. Like the master craftsmen who dotted Europe with gothic cathedrals, I started to get involved in the making as much as the drawing. In fact, at some point, I made pieces which I hadn’t drawn yet. During machining, I became even more aware of the “bend don’t break” mantra. On CAD software, architects draw straight lines and perfect perpendicular edges. In reality, nothing is straight. If you disagree, use a micrometer. Tolerances, flexible mountings and rubber spacers accommodate these “errors” and are more effective than high precision manufacturing. So why do we speak of 3D printed buildings and CNC fabrication? After all, this is architecture not medicine. To be honest, I don’t know. I think the way forward is not to reduce tolerance but to increase it.

Submit a Comment