When Buildings Fail

Architectural malfunctions have as much to do with the culture of design as they do with faulty construction.

In 2005, just months after it opened, gym locker-sized pieces of glass started falling off Montreal’s shiny new Grande Bibliothèque, shattering on the street below. A building had failed. But at least no one was hurt.

As a quick fix, authorities put up a fence around the $100 million library. But they found it difficult to address the underlying cause. That’s because building failures are not purely technical problems, but cultural ones as well.

Recently we’ve seen a spate of high-profile failures, including malfunctioning elevators in the world’s tallest skyscraper, the Burj Khalifa in Dubai, and the leaky roof of Daniel Libeskind’s exuberant Denver Art Museum. These are innocuous, if sometimes expensive, examples. “Blame” here just means finding out who’s going to pay for the repairs – typically by suing someone. Yet failure can also bring tragedy. Last year in Montreal, a piece of falling concrete killed a woman seated in a sidewalk sushi restaurant.

Who’s to blame when buildings fail?

Cost-conscious clients perhaps. The Bibliothèque’s designers, a consortium of Vancouver-based Patkau Architects and Quebec City-based Croft Pelletier architects, got the job through a widely publicized international competition. (A Montreal firm, Menkès Shooner Dagenais, joined the team after the competition.) Jurors praised their design’s sophisticated copper-covered façades. Alas, the copper was quickly ditched as too expensive. So the architects invented a new cladding composed of 6,200 “C”-shaped planks of channel glass, tempered and tinted with a green ceramic coating. But why didn’t the library simply pony up the cash to build the copper prizewinner they commissioned?

We can also blame those architects who, in their search for novelty, push the limits of building technology. The Bibliothèque’s architects did indeed innovate. Such glass planks are typically installed vertically, as Steven Holl had used them around the same time for the Nelson-Atkins Museum of Art in Kansas. The Patkau team opted to position the planks horizontally, an installation never before seen in North America.

And then there’s politics. Before the library opened, the directors bragged that the planks were to be made in Quebec. That is, rather than use a standard product from an international manufacturer, they wanted to build with a Quebec product. Once the glass started to fall, however, this hometown pride disappeared from the library’s press releases.

The blame game never ends. Perhaps the contractors didn’t build according to specifications. Accidents, sloppy work, and mistakes can mean that even buildings constructed in standard ways with standard materials fail. Tests revealed that up to 30 per cent of the 6,200 planks did not meet Canadian norms for safety glass. Perhaps the manufacturer didn’t make the product properly.

But buildings are intricate social products, and we have to understand their failures accordingly.

To get at this, we need to consider the very nature of design. First of all, there are good reasons for architectural experimentation beyond artistic freedom. Some innovations work extremely well – so well they become the new standard way of doing things. New technologies, such as ways to assure sustainable architecture, are highly desirable.

Second, failure is part of how design proceeds, whether we’re making iPods or Toyotas. In architecture, the reason you hire an architect in the first place rather than just remake the same building over and over is because you want a customized, unique design. In effect, each piece of architecture is a prototype, an experiment. That’s how design works.

The Bibliothèque’s final solution? The library replaced individual 2.5 metre-long planks as they fell for about $3,000 each. But administrators balked at paying the $3 million it would cost to replace them all. Instead, the architects were asked to design a $750,000 “security perimeter.” Dense shrubs now keep passersby away from the building and aluminum canopies shield visitors at the entrances. It’s a strange “solution,” since it doesn’t fix the problem and it hides the building’s design behind shrubbery.

Yet the so-called solution at least has this value: that the main actors involved collaborated to resolve the problem and didn’t just blame each other. Finding someone to blame is often just the way we express our anxiety when technology fails. Instead of blame, we need this kind of negotiation. We need to understand that hospitals, dams, bridges, and libraries are made out of social relationships as much as they are built with concrete and glass.

In the end, we have to understand that building failures reveal as much about us as building successes.