Fictions etched by rain in rock

If modernism was haunted by time, postmodernism is obsessed by space. Time in a postmodern age no longer means decay, duration, ennui, but the international clocks of the Stock Exchange, where different chronologies are stacked side by side in the same location. In an older form of capitalism, it was time that was money; now it is space that is costly, largely in the shape of real estate. The media and the transnational corporations have reconfigured global space, dramatically shrinking distance and bringing Hong Kong closer to London than Halifax. On a map of market accessibility to the European "core", Belfast is fairly close to Bucharest. But if the great global corporations erode distances, they also despoil environments, thus putting another kind of spatial preoccupation - ecology - on the postmodern agenda.

For some postmodernists in Ivy League universities, as for some apocalypticists holed up in bunkers in Montana, time in the form of history is now definitively over. We live in a windless enclosure in which the same elements are ceaselessly shuffled into different patterns, a space which could be ruptured by nothing as absurdly original as an historic event. History in the sense of the past is converted into a consumable commodity known as "heritage", while the future has been called off for lack of interest. No doubt it really would take the second coming of Christ to disrupt this bland continuum, though even this is questionable: one hears that certain US evangelical television stations are already working on the best camera angles to record the spectacle.

Like Tamberlaine or Alexander, modern technology eats up great tracts of space, and is thus always at risk of hubris. Time, by contrast, chastens by recalling us to our mortality. But time is also the medium of human overreaching, while the spatial in the shape of the body reminds us of our material limits. Technologies that can span space and time spring from human bodies that are subject to the constraints of both - stubbornly local entities whose very materiality makes them vulnerable. It is only when we open our mouths to speak that we reveal how much we are, in principle at least, universal creatures.

It seems inevitable, then, that a literary criticism that loudly proclaims its materialism was finally going to rediscover that most materialist of all intellectual disciplines, geography. For here is materialism not just in the form of sexuality or the signifier, but in the shape of rainfall and rock formations, of culture as coded and moulded by its physical environment. What we now know as the sociology of literature began in 18th-century Europe as a kind of topographical discourse, all to do with the relations between terrain, climate, customs, morals, temperament, art and the like. Somewhat to the embarrassment of today's postmodernist devotees of the spatial, it was also a form of Enlightenment rationalism. And since geography and literary criticism are both marginal, mongrelised discourses of distressingly low definition - great portmanteau-like subjects which range in the former case from sand dunes to shopping precincts and in the latter case from dactyls to death - their eventual convergence was all the more probable.

Geography, however, can question postmodern assumptions as much as underpin them. Maps, for example, make connections, whereas postmodernism prefers on the whole to highlight differences; they can be global in scope, in contrast to postmodernism's dogmatic anti-universalism; and though they can be construed in all kinds of ways, which fits a postmodern epistemology, they are evidently representations of a world beyond themselves, which chimes rather more with epistemological realism. Franco Moretti's Atlas of the European Novel is crammed with fascinating literary maps, which chart such things as the movements of four different Dickens heroes, the diffusion of translations of Don Quixote throughout 19th-century Europe, the geographical spots where Dickens's novels end, and those in which Jane Austen's narratives tend to become especially complicated. There is even a map displaying the countries of origin of 19th-century fictional villains. The English reader will be gratified to learn that "France is clearly the epicentre of the world's evils", and that all the wrong male sexual choices of the English Bildungsroman involve women with French connections.

Moretti is one of the most adventurous of contemporary critics, keen to engage in ambitious speculation in a climate cautious of structural generalities. The bold conjectures of two of his previous works, The Way of the World and Modern Epic , have transformed our view of the European novel, and this latest offering seems set to do the same. Hardly any English critic can match his comparativist's range of reference, and few can equal the amicable luicidity of his prose. Moretti is a specialist in the two or even one-word sentence and can sustain grammatically impeccable paragraphs with scarcely a main verb in sight.

Literary geography, he reminds us, can refer to the study of space in literature, or to the study of literature in space. This book engages in both kinds of inquiry, alert to the "place-bound nature of literary forms", but equally insistent on how "geography shapes the narrative structure of the European novel". In modern Europe, that new-fangled, impalpable space called the nation needed its appropriate symbolic representation, and the totalising, comprehensive novel form was at hand to provide it. Moretti is adept at showing how novelistic narrative charts spatial unities and divisions, throwing in some intriguing research findings in the process: historical novels are usually located somewhere near natural barriers such as forests, mountains and coastlines, while rhetoric and metaphor tend to thrive close to borders, in uncharted terrains which only figurative prose can handle. Progress in the Spanish picaresque novel depends on the amblings of the mule, a suitably unheroic mode for such a prosaic, everyday genre.

The strangest aspect of such bizarre speculations is just how persuasive they are. There are some finely detailed explorations of Balzac's Paris and Dickens's London, along with an erudite chapter on "Narrative markets", which plots literary production in 19th-century Europe on an import/export model. England and France were major exporters of novelistic genres, while other nations either imported them or - in what modern economists would call "import substitution" - fashioned their own local imitations.

The book, however, is more than a sociology of literature. Its achievement is not just to set novels in physical and social space, but to demonstrate how these spatial coordinates shape their plot and even, occasionally, their style. Moretti's "scientific" bent as a critic, fascinated as he is by patterns, regularities, remote affinities, will send most English critics scrambling defensively for their unique particularities. But here, as in his earlier studies, these uncompromisingly "continental" methods serve to illuminate fictional details in estrangingly original ways - and this, after all, is just what the "words-on-the-page" merchants have been clamouring for.

Terry Eagleton is professor of English literature, University of Oxford.