The Faber Book of Science, edited by John Carey (Faber) £17.50

The world is a fluent place. Physicists borrow quarks from Finnegans Wake (itself an encyclopedia of correspondences) and lend allusions of strangeness and flavour to particle behaviour: there are crashing computers and left-handed neutrinos, the selfish gene and parasitic DNA, and colourfully violent comic-strip imagery in cosmology: and stranger and stranger still, on the other side of the room deconstructionists and comparative linguists adopt algebraic protocols, literary critics yield to indeterminacy and social scientists acquire expertise in confidence intervals.

A mere thirty years ago this grotesque pangolin of artist engineer imagery would have been shot dead on hearsay, and its carcass quarantined by scare quotes. Nowadays its ubiquity scarcely raises an eyebrow.

Our society may be more adept than it knows at living with conflicting epistemologies: the history of technology is also a history of language, for the word technology itself, when first used in the time of Liebniz's De Arte Combinatoria and the universal language projects of Dalgamo and Wilkins, was a term applied uniquely to grammar - to the 'science of rhetoric'. Three centuries later we are casually aware of how that formal syntactic manipulation of experience undermines the very need it determines - for the concrete. As usual, Walter Benjamin noticed long before anyone else. In a letter to Gershom Scholem, he cited this passage from Eddington's The Nature of the Physical World:

The plank has no ...