Penguins and causality

Architect Berthold Lubetkin and engineers Ove Arup and Felix Samuely created the penguin pool at London Zoo, completed in 1934. It’s a great example of modernist architecture.

The main feature is a pair of intertwining ramps for the penguins which appear to rise unsupported from the pool. The concrete ramps are very thin, and this structure is only possible because of the use of reinforced concrete in its construction:

penguinPool_295x182_Arup

Our structure of established causal claims is analogous. We usually have a limited amount of evidence of correlations (concrete), but by strengthening it with evidence of mechanisms (the steel mesh in reinforced concrete), total evidence is strong enough to support the structure as a whole. The art is to achieve a balance between evidence of mechanisms and the evidence of correlations so that costly data collection can be kept to a minimum while retaining sufficient strength to support the structure. The result can be not only elegant but also enduring.

penguin-pool-lubetkins-archeyes-12

 

Of course, a structure of causal claims can be as elegant as you like, but it has to lead to somewhere desirable. Unfortunately, the water at the end of the ramps wasn’t deep enough for the penguins and they eventually had to move to a duck pond.

Penguin Pool, London Zoo, Regent's Park, London: model of the pool with penguins
Model of the pool with penguins

There’s a nice exhibition on Ove Arup’s work at the V&A museum in London at the moment.

Leave a Reply

Your email address will not be published. Required fields are marked *