BOOK REVIEW / Landscapes composed by the inner eye: Fields of Vision - Stephen Daniels: Polity Press, pounds 12.95

GEOGRAPHY? Aren't we pretty sure that it's the dull recounting of rivers deep and mountains high? Here is a book that bids fair to inspire the most rooted sceptic. It is elegantly written (by a senior lecturer at the University of Nottingham) and is a lively exploration of the way landscape has been discussed and portrayed in art.

The book includes accounts of two famous exhibitions: one of American 'Western Art', the other of Constable. These shows argued that the paintings purveyed myths (of frontier nobility and rural innocence, respectively). There is no 'innocent' landscape, we gather, and no innocent landscape art. We go out to the natural world, or its near-cousin, the farmscape, to find something that pre-dates man. We go out to find the Great Other.

But we also go out with eyes and minds aleady buzzing, and do not so much find things in the landscape as put them there. The more we see paintings of the natural world, the more this is driven home, and never more clearly than when paintings are 'packaged' in galleries and catalogues. Then, we pick up the meanings that other people have invested them with.

There is a fairly familiar discussion in the book about how modern man has come to love, instead of fear, nature and wilderness. For most of us, the master text here is Keith Thomas's Man and the Natural World. Daniels shows that people always did have ambiguous feelings about nature. Man has not, as some green-minded people imagine, gone from a homogeneous unfeeling dark age to a new enlightenment.

In his essay on St Paul's (well worth reading, though whether architecture can be landscape bears a bit of thought), we see that at different times, and to different people, the dome has seemed an expression of state power, and at other times and to other people, a populist symbol. Daniels does not argue the toss one way or the other: instead we have a Thomas- like delight in the texts surrounding the issue, and in their contradictoriness.

It perhaps goes without saying that Daniels's human geography sometimes looks a litle like sociology. But Daniels doesn't show a common sociologist's knowingness, and neither does he flaunt anti-establishment feelings. He seems to take the evidence as it comes and treats even the privileged with a degree of respect (this matters in discussions which must involve landowners and patronage).

Some of the byways he explores are illuminating. It transpires, in an essay on Joseph Wright, that 18th-century tourism included much nervous excitement about one of the obvious bogeymen of the age: industrial engineering. The enthusiasm of polite people for factory-viewing 200 years ago (whatever Wordsworth might say) makes a nice counterpoint to the success of Sellafield as a tourist attraction now (whatever Greenpeace might say).

This is the nub of the book. Artists (the book discusses J M W Turner here, and Thomas Cole in the US) and landscapers such as Humphrey Repton seemed to have shown themselves suitably ambivalent about the forces of progress surrounding them. They were thrilled and dismayed in equal doses, and produced images that can be read either way.

Similarly, people have used different bits of actual landscape, and different art works, as icons of a national identity. But this was never certain or fixed. Sometimes the British have felt themselves to be progressive, at others archaic; sometimes urban, sometimes rustic; either assertive or reserved. It seems fitting that this book should appear to remind us that our landscape, like our identity, has always been in a state of flux.