Mark Hemel and Barbara Kuit were educated at Delft University of Technology and opened their office in 1998 in London. Barbara worked for Zaha Hadid and Mark taught at the Architectural Association. Under the name Information Based Architecture they entered big competitions and won a number of prizes, although none of these submissions resulted in a building commission. In 2003 they relocated to Amsterdam. At that time the office was working on a competition for a Sightseeing and TV Tower in Guangzhou. One day they received a phone call informing them that their submission was among the final three candidates.

So time for champagne, since your design was a contender for construction?
Well... we were also concerned. We weren’t exactly in a party mood. We were more focused on doing it to the best of our ability.

How was it at the start? You were a small office and then suddenly such a big commission.

At the time we had about thirteen people in the office. When working on a competition we always formed a bigger team. But there was so much happening back then. We had to decide whether or not to move to China, even though we had just had a baby. What’s more, it was not very clear for a long time whether we really had won the competition or not. The competition ran from April to August 2004. Then the client spent a long time negotiating with three teams. It was only in February 2005 that it became clear that they were serious about working with us. Then the contract negotiations with all parties could start. That was very exciting. We had to find out what the risks were, all the things we had to consider.
The ground-breaking ceremony took place in November 2005. At the start we mostly worked on the lower part, which also included a metro station, even though the tower to be built above was not yet fully detailed. Digging and constructing the foundations alone took a year, and the tower had to be completed by 2010, because that’s when the Asian Games were to be held in Guangzhou. So the builders had to get going. But working out the lower part first meant we couldn’t work on the whole design and so we weren’t able to keep an eye on the whole design.

But you had come up with an incredibly strong image that couldn’t be reduced. Otherwise there’d be nothing left.
That’s why we deliberately incorporated a measure of flexibility. You could reduce the rotation a little, which would immediately reduce costs significantly. The greater the twist, the narrower the centre and the more steel required. So everyone involved was busy reducing the amount of twist, except the client.

What changed during the construction test phase?
Very little, thankfully. In the end the tower became a little wider, and the vertical steel ‘lines’ are filled with concrete. What’s more, ‘mass dempers’ were added at a later stage. They take the form of two big water tanks on rollers to counter the ‘sway’ of the tower. The design consists of a hyperbolic form, and that was structurally advantageous. Such a choice came about very intuitively. During the competition you imagine a shape you believe in; and then everyone keeps on questioning it. But in the end it was okay. Making the tower narrower would reduce the wind pressure. You cannot calculate all that many things very precisely in advance. That’s the nice thing, since technology is decisive in projects like this, and you have to discover a lot of things through tests; that leaves a lot of scope for intuition.

That sounds absolutely perfect, but did anything do wrong? Things that you later look back on and think you should have done differently?

Some things did go wrong, but you have to move on. Our office is only one small partner in the whole process. The design team, with Arup as an important partner right from an early stage, was augmented with numerous engineers and consultants immediately after the competition. The client also assembled a similar team. All sorts of experts were flown in every time to check what we’d done. We were a lone operator among all those different parties. Most of them spoke the language of functionalism and money. That’s where they could score points. Architects, by contrast, talk about vague things – things they would be ‘sorry’ to see scrapped. Yet in hindsight we think they listened more to us more than we thought at the time. For example, they wanted the rings, now positioned at an angle, placed horizontally because it would have been cheaper. Then we explained that would mean looking at one of the rings every few floors, while if they were at an angle you could look over and beneath them.

When you see this design you think there must be an extremely good calculation behind it. And it’s no coincidence that your name is Information Based Architecture. How do you work? What do you aim for?
In contemporary architecture you see lots of forms that refer to nothing. And indeed, we want to make designs that are more 'informed'. At the same time, it could be anything. It’s a sort of biological process; you plant a seed that contains an awful lot of information. That seed then grows into something that is not determined in advance. That’s why we try to make our models dynamic. With the tower we started with wire models, which are actually similar to computer models, but then made physical. Just like those suspended models that Gaudi made. They already contain a structural element and, at the same time, they reveal something spatial. It’s a coherent system, while modernism means splitting up everything – the structure separate from the façade, the ventilation strips separate from the windows and so on. We try to bring everything together in one single solution. The difficult thing with that is that everything influences everything else. But that’s precisely the interesting thing too; since every change can influence the whole thing, you can’t do everything and you guide the design. And things emerge that you could not have predicted beforehand.

How did the form of the Canton Tower emerge?
We started by making models. The lines are connected by forces that are transferred to the ground. That is imperative in such a tower, so the lines are there from the start. They are very long, continuous lines to keep everything as simple as possible. At the same time, we didn’t want to make a very straight object with those straight lines but, rather, an almost fluid design. To us that has to do with the times we live in, with a possibility …

...of a sort of aesthetic translation of the zeitgeist?
Exactly. But the twisting also has to do with the location in the city. They are constructing an almost Parisian axis perpendicular to the river. We felt that the tower should relate to both the axis and the river. The rotation ensures that. All those elements influence the design. It’s very strange. Intuitively you head in a particular direction and then it turns out that if you have to formulate and explain why you made the choices you did, then it turns out to work at different levels.

The provisional subtitle of the book about your work from 010 Publishers is 'biological intelligence'. Is that an upgrade from the Information Based in your office name? You can still interpret 'information' as a form of functionalism, but if it’s biological, then you relinquish some control to an evolutionary development of nature. Is that a conscious choice?
That’s definitely a conscious choice. What we’re looking for has to do with processes; you use forces and structures because those are components that every building possesses. If you take them as the starting point for your design, then you immediately have a better grip on the building. The initial information is often strong and unequivocal, like DNA, and can steer the design in all sorts of directions, but it still remains recognisable and understandable. It’s the same with food. The best food is made with a limited and carefully considered number of ingredients, so that you can still taste what went into the recipe. Such a start could be a line. Once you have that line, you can start thinking about the thickness of the line and you can vary the angle of the line. You can come up with a huge number of variations using just a few parameters. For instance, in the tower we made the column 2 metres thick at the bottom, narrowing to 1.1 metre at the top. That may seem a very minor variation, but if the rotation then brings the lines closer together and further apart towards the top, you experience that you are entering a tall tower all the more.