Flexible buildings: the problem with mixing layers

Different parts of a building evolve at different speeds, so it makes sense to keep them separate

June 2021

Words by Tony Whitehead

“Services technology is likely to require updating more quickly than the structure, which may never change significantly. The idea is that if you keep these layers separate, they can evolve at different speeds and be adapted relatively easily”

Robert Schmidt III, Loughborough University

How do buildings adapt? How should they adapt? These questions have provoked some of the deepest thinking in architectural academia, and the books on these subjects are less about light, space and proportion than time, society and life.

Christopher Wray’s The Timeless Way of Building laments that too much design has become a top-down process, with a damaging distance between designer and user. Beautiful towns and buildings, he argues, are arrived at via thousands of acts of individual creativity over time — so buildings need to last a long time if only to give the users time to “tamper” with them. Following this logic, it might take decades or even centuries before they reach an optimally adapted state. 

The ultimate flexible building

Read the full article

This is not to say that buildings cannot be designed to adapt. Architect Frank Duffy has argued that they comprise several layers — shell, services, fittings and contents — all of which evolve over different timescales. The idea of these “shearing layers” was developed further by Stewart Brand in his 1994 book How Buildings Learn: What Happens After They’re Built. (Answer: even if they are not designed to adapt, they all adapt anyway.)

Robert Schmidt III, reader in architecture at Loughborough University and co-author of Adaptable Architecture, believes that shearing layers is a useful concept. “Services technology, for example, is likely to require updating more quickly than the structure, which may never change significantly. The idea is that if you keep these layers separate, they can evolve at different speeds and be adapted relatively easily.”

Using this principle, he says, a facade that is easily detachable from the structure becomes more readily adaptable than one that isn’t. “But this way of thinking can run contrary to some of the current ideas about making every part of a building work hard for you by introducing multifunctional elements. You might look to integrate some services into the facade, but that can make upgrading services or adapting the facade more difficult. Similarly, if you integrate PV solar into your roof, it’s going to make updating the solar or the roof much harder.”

In contrast, functions that have been usefully combined for millennia are unlikely to compromise adaptability. Windows combine light and ventilation in a way that seldom goes out of fashion, and a structure whose thermal mass also regulates temperature can retain the function for the life of the building irrespective of changing use or new technology. “One reason old buildings adapt well is because technology used to evolve more slowly,” says Schmidt. “In many ways adaptability is supported by simplicity and familiarity.”

This article appears in The Possible issue 07, as part of a longer feature on flexible buildings

Pirelli Building
The Pirelli Tyre Building in New Haven, Connecticut, US, by Marcel Breuer (1970). Brutalist buildings are not usually perceived to be flexible, but this former office building had two factors in its favour: a celebrated design with geometric detailing and a dramatic mid-level void, which helped heritage groups to defend its preservation; and a robust, durable structure that could be detached from shorter-lived layers such as services and fittings. It has now been reimagined as a net-zero energy hotel by Becker + Becker
Photo Randy Duchaine / Alamy Stock Photo

Leave a comment

Leave a Comment

Your email address will not be published.