Mud, spit and flights of fancy

Animal Architects: Building and the Evolution of Intelligence By James R. Gould and Carol Grant Gould Basic Books, 2pp, £15.99 ISBN 97804650828

Nicola Clayton enters the extraordinary world of animal architecture

We tend to think of ourselves as the world's master builders because of the complexity with which humans plan, design and construct buildings, not to mention our ability to create cities and cultivate landscapes. Yet the ability to build did not evolve de novo but is something we have in common with a number of other animals. In fact, most animal builders belong to one of just three classes of animals, the Arachnida (spiders and mites), the Insecta (insects, including termites) and Aves (birds). Within these groups, there is considerable variety in the size, shape and type of structures built, from the highly elaborate trap nets made by Deinopis subrufa , the net-casting spider, which holds its net with its four front legs, waiting to ensnare the next victim, to the complex nests built by weaver birds. Some of them are works of construction. Consider the saliva-laden soil-termite nests of the Australian termite, Nasutitermes triodiae , that stand up to 7.5m tall, assembled by animals that are just 0.25cm long. By human standards, the equivalent would be a skyscraper 10km high. And like many of our modern skyscrapers, these termite towers contain a sophisticated air-conditioning system. In other ways these nests are more like a palace with a royal chamber that houses the king and queen, along with nurseries, feeding chambers, bathrooms and gardens complete with a well.

There have been a number of books written on animal architecture and the biology of building behaviour, but in Animal Architects: Building and the Evolution of Intelligence , James R. Gould and Carol Grant Gould beautifully describe some of the architectural wonders of the animal kingdom. They start by addressing the question of why animals build and explaining in a way understandable to the layperson that some building behaviours appear to follow a fixed motor programme, whereas others may be much more flexible and involve considerable learning (what they term "learning to recognise" versus "learning to do"). The Goulds then go on to describe animal architecture across a range of species, starting with those that seem to be hard wired (such as the nests of hunting wasps) to those whose structures may require more cognitive feats, with a detailed discussion of the "cognitive map" and how it may have evolved.

Their goal is twofold. The first is to inform the general reader about the diversity of animal artefacts, and how and why these structures are made. Their second goal is more ambitious, namely to investigate the role of planning and decision-making, as well as instinct and learning, to ask what, if anything, an understanding of the biology of animal building can tell us about the evolution of intelligence. They argue that the structures that animals build provide a wonderful window from which to observe the way in which the animal mind works, and the evolution and nature of our own intelligence.

The book is certainly an interesting read, and there is no doubt that the Goulds succeed in their first goal of captivating the reader with their enthusiasm and encyclopaedic knowledge of the biology of building. Unfortunately, they do not relate animal construction to human architecture other than a rather cursory section in the final chapter on possible scenarios of the evolution of the human mind. This surprised me because, according to the very earliest surviving work on human architecture, Vitruvius's De architectura , good buildings satisfy three core principles: firmness, commodity and delight - in other words, structure, function and aesthetics. So one could ask to what extent these principles may also be applied to animal architecture. In terms of structure, animals have evolved many architectural features with which we are familiar, from the air-conditioning systems and wells developed by the termites to the hinged doors with handles used by trapdoor spiders and the roadways built by ants. Clearly most examples of animal architecture are entirely functionalist: animals are uncompromised by the need to represent their architecture symbolically so, unlike us, it is not dictated by cultural constraints, sociopolitical or aesthetic aims. The philosopher Karsten Harries wrote: "Architecture is not only about domesticating space, it is also a deep defence against the terror of time. The language of beauty is essentially the language of timeless reality." No wonder, then, that Sverre Fehn, the architect who won the Pritzker Prize in 1997, referred to the bird's nest as "absolute functionalism, because the bird is not aware of its death".

But what of aesthetics? Le Corbusier said: "You employ stone, wood and concrete, and with these materials you build houses and palaces: that is construction. Ingenuity is at work. But suddenly you touch my heart, you do me good. I am happy and I say: 'This is beautiful. That is architecture.' Art enters in..." Some animal buildings do meet this criterion, and the aesthetic beauty of the bowers built by the various species of bowerbird are described in delightful detail in this book. These bowers, while thatched with sticks, are not nests but art galleries. In other words, bowers are not structures built to enhance the bird's own survival directly, so in this sense they are not purely functionalist. Instead they are designed by males to attract the females with their artistic skills. These bowers are evaluated in great detail by the females, who choose whether or not to mate with the gallery holder based on his artistic prowess, and also by rival males, who will readily steal from his collection should the opportunity arise. In terms of pure aesthetics, I would argue that it is hard to beat the intricately and colourfully decorated maypole bowers of Prionodura newtoniania , the golden bowerbird. These birds make twin maypoles, constructing columns of sticks around the stem of a sapling or small fern, which is bridged by a trailing vine stem that the males use as a display perch. They decorate the bower with lichens, berries and even flower arrangements, all with the aim of wooing the ladies. What these examples show is that aesthetics can enter into animal architecture through the process of sexual selection, a point that the Goulds discuss in this, their most interesting chapter. They argue that sexual selection has also shaped the evolution of our own minds.

The Gould's second goal was to investigate the role of cognitive processes and to assess the extent to which an understanding of the biology of animal building can tell us about the evolution of intelligence. As a comparative psychologist, I found this goal more ambitious and the outcome less successful. I was disappointed not to find a more detailed discussion, given the subtitle "Building and the evolution of intelligence". For one thing, I would like to have seen a more substantial analysis in terms of what possible cognitive processes might be involved and how such processes might contribute to animal architecture and to the evolution of intelligence. The authors mention insight, folk physics and planning, yet I felt the issues were not unpacked in any satisfactory detail. The section on folk physics, for example, did not describe any of the seminal work by Danny Povinelli and his colleagues on the apparent lack of causal understanding by chimps, nor the careful studies by Elisabetta Visalberghi and her colleagues on how primates perform on various tests of physical intelligence, which have sparked a suite of comparative studies on other animals, including finches and crows. Nor could I decipher a clear explanation of what the Goulds meant by "planning", other than a discourse on cognitive maps. In comparative psychology, planning usually refers to forethought: the ability to take action now for a future need independent of the animal's current needs. In addition, this should be a novel action or combination of actions to rule out responses that are triggered by seasonal cues such as migration and hibernation, neither of which may require any planning, and responses that are learnt through trial and error, what the Goulds call "learning to do". So perhaps it is no wonder that, until recently, this ability was thought to be unique to humans.

What I did like was the authors' emphasis on "looking at ways in which the challenges of a species' niche are often a better predictor of mental abilities than phylogeny". In other words, we should avoid being primatocentric by default. Indeed, one of the most convincing examples of animal architecture, at least according to Le Corbusier's definition, comes not from our close relatives, the chimpanzees, but from the bowerbirds.

Nicola Clayton is professor of comparative cognition, Cambridge University.