Stars in their eyes

Before this book is wrongly shelved in the animal section, it should be made clear that the subject matter is human cognition and its relationship to culture. The "Wild" part of the title refers to the anthropological approach of author Ed Hutchins, who studies naturally situated cognition in the real world, rather than in an artificial laboratory environment. As a racing yacht navigator, he chose navigation as the activity to study, and did the fieldwork for the book aboard United States Navy ships observing the navigation team. He develops the thesis that the cognitive skills used are socially distributed: both among the team on the navigation bridge, and further within the use of instruments, charts, and conventions that are themselves all social practices.

For fish, the presence of water is not an issue, except under dire circumstances, and there is nothing so difficult to see as our own cultural background. The closer one is to a subject, the more difficult it is to achieve the objective understanding that our own 20th-century western scientific culture demands. So it is just as well that the technical details of navigational practices will be unfamiliar to many readers. These are spelt out in some detail in the context of the choreographed routines of the navigation team, which is occasionally supplemented by improvisation in a crisis.

Further distance in space and time is given by comparing modern western navigation with that of other cultures and of other eras. There is a delightful discussion of how the Micronesians can embark on ocean voyages in outrigger canoes out of sight of land for several days, and arrive at a small island without any instruments. Their primary frame of reference is the stars and the paths of the constellations across the sky. These locate a fixed frame centred on the observer in his canoe - hence it is the islands that move relative to this frame. The navigator sets sail in the right direction, and as the water rushes by the boat it takes along with it memorised reference islands out on one side or the other, until the destination hoves into view. The reference points, which are passed on orally in this nonliterate culture, are out of sight on the beam; by estimating their speed from the flow of water, and judging elapsed time, the navigator keeps track in his mind's eye of where all the islands are.

This method works - it "stands the stern test of landfall". Yet to western preconceptions of islands with a fixed position on a chart, and a moving ship, it seems paradoxical for the Micronesians to claim that the islands move around a stationary vessel. Hutchins reconciles these views through David Marr's notions of levels of analysis of cognitive systems: the same underlying computational constraints of the navigation task can be satisfied by a wide choice of algorithmic and representational schemes, contingent on historical processes by which effective practices are discovered, codified and passed on within a particular culture.

Hutchins is concerned to relocate the boundaries of cognition, away from the skull of an individual into the practices of a social group. This socially distributed cognition can be seen explicitly in the coordinated teamwork on board the US naval ship; but even a solo navigator is dependent on conventions which represent the accumulated practical wisdom of her culture.

The book concludes with a critique of "official" cognitive science, and its fellow-traveller GOFAI (Good Old Fashioned Artificial Intelligence), which has assumed that cognition is to do with individuals manipulating symbolic representations inside their head. Hutchins joins the swelling group of people who realise that the computational metaphor for the mind has passed its sell-by date. A model of human intelligence as disembodied and unsituated omits the sensorimotor processes and cultural conventions which give it context - in fact omits its humanity; by studying cognition in the wild Hutchins draws this back in. I was sorry to miss a further comparison with animal navigation methods, which could have drawn attention to our animality; but the gap between us cultural humans and the beasts is more difficult to straddle than that between us "rational" westerners and those exotic other cultures with their "strange ideas".

I came away from this book realising how potent is the metaphor of a chart for the scientific obsession with objectivity. The ocean and the ship containing us is viewed from above by an external unsituated navigator. The Micronesian navigational framework, centred on the observer, is an equally potent metaphor for a relativistic view of our world.

Inman Harvey is research fellow in evolutionary robotics, school of cognitive and computing sciences, University of Sussex.