Everybody's Dear

The intentional fallacy is the New Critical tenet that has proved most enduringly cogent. That the enthusiasm for literary biography is undiminished ought to be regrettable: lives priced at over a shilling are liable to give us rather more than the facts. Yet critical practice and wisdom maintain that any interpretation based on biographical data can be trumped by a close reading of the text itself. Such have been the rules of scholarly engagement, the unifying premise of the most diverse and antagonistic schools of literary criticism, whatever their 'theoretical' claims or alignments.

Over the past twenty years we have seen the words 'life' and 'biography' extended into the inorganic world of paper. This is a consequence of the welcome resurgence of interest in bibliography, an interest that has thoroughly unsettled any notion of 'the text itself'. Close reading has been exposed as naive: its disdain for 'context' – and its assumption that text can be distinguished from context – is shown to be unwarranted: as a stable and independent object of study, the text has at last proved elusive. It is as though five hundred years had to pass before it could be acknowledged that print was not fixity. 'Print settles it', wrote – to be printed – Charles Lamb, lamenting the survival of a poet's handwriting, even, or especially, the manuscript of 'Lycidas'.

Looking at the text used to be a discipline of focus; now peripheral vision is required. The obvious (the designedly self-adverting) attractions of the paratext had evaded all but episodic or anecdotal attention, beyond specialised bibliographical scrutiny, of cancels, or paste-downs. The publication of Gérard Genette's Seuils in 1987, translated into English as Paratexts in 1997, has proved decisive for critical reading. Genette's concern with what goes on 'around the text', the prefatory matter, title, illustrations, colophons, indices, dedications, page numbers, running headers, not to overlook all manner of promotional contiguities, has enormously enriched our sense of what, in a book, matters: above all, that the text exists in matter, that there is no ideal 'disembodied' or merely contingently embodied text: that without a paratext, no text is, nor can be, read.

Equally influential has been the revolution in bibliographical concerns inspired by D. F. McKenzie. His work has given a theoretical edge to historical criticism, and has provided much support to those who would hold that theory cannot be purged of history. Where Genette points to the spatial improprieties, the porous discontents of the text, McKenzie's concept of the 'sociology of texts' insists that no writing can have temporal fixity: a text bears the marks of its making, and contains those of its earlier readings, its precedent printings. McKenzie's 'Bibliography and the Sociology of Texts' was first printed in 1986 by the British Library as the Panizzi Lectures; only in 1999, after its estimable success in French translation, was the book 'published' (by Cambridge University Press). McKenzie resolves the dilemma that confronts any critical reader: whether to derive a text's meaning from the authorial context, by diachronical analysis, or to remain within the theoretically legitimate synchronic, to find meaning in the present act of reading. He has directed our material attention to the possibilities of understanding the process of mediation: it can no longer be supposed that the text we hold in our hands could have a meaning independent of the process whereby it arrived there. Kathryn Sutherland cites McKenzie once (p. 36 n. 63), but his name has most unjustly slipped the index.

Twenty or thirty years ago scholars might berate critics for citing the works of, say, Jane Austen in, say, the Penguin edition. The rule was that one should cite either the first edition, or the last edition seen to press by the author, or the modern scholarly edition whose text had been established in a way that could be deemed 'authoritative'. While the New Criticism denied the materiality of the text, the old bibliography made a fetish out of almost any text except the ones likely to...