Abstractions of Evidence in the Study of Manuscripts and Early Printed Books (review)

Joseph Dane's The Myth of Print Culture (2003) asked us to think about a major problem of critical practice in bibliography: the relationship between the evidence studied by scholars of the book and the framing assumptions they necessarily bring to bear on those books in order to study them. Forcefully and with considerable wit he argued that 'the gap between material and textual levels in bibliographical discussion is one that can never be closed, and it is one that scholarship, in its own advance, discovers new and more mystifying ways to obscure' (pp. 3–4). Nominalist in argument, The Myth of Print Culture was also nominalist in form: a book of sepa rate essays, profoundly and usually amusingly sceptical. Reviewing it (in The Library, vii, 5 (2004), 322–25) I wrote that 'if we accept Dane's assertion that philosophical principles are at issue […] we are entitled to ask if the principles he asserts are adequate. […] Dane has difficulty admitting the power of hypothesis to search out truth, because he seems to assume that all hypotheses, and with them all general explanations, are kinds of absolutism.'

Abstractions of Evidence in the Study of Manuscripts and Early Printed Books is a shorter book, and less obviously polemical (it also has more typos), but once again it is a book of essays, and Dane's resistance to general explanations is unaltered. Indeed, both Introduction and Conclusion suggest that the very concept of evidence itself might be an abstraction. In the Rare Book Room, for example, the visiting scholar may be handed 'an object that is not quite what was envisioned: a photo facsimile in a dull green cover, a note from a librarian, the word reliure or one of its many variants, an empty desk, a reference to EEBO, the wrong book' (p. 1). Then there is the catalogue itself, which refers to objects that are 'not actual books, but rather ideas or descriptions of books' (p. 3); the catalogue is a generalization, subject to the constant degradation inherent in the frail historical process of bibliographical description. The succeeding essays focus on how that act of abstraction comes about in particular cases of bibliographical description.

In Part 1, 'Inference and Evidence in Medieval Books', the first chapter (written with Rosemary A. Roberts) applies probability theory to Greg's account (1910) of the ten possible editions of Everyman. The equations in this essay are beyond me, but its conclusions are not: the mathematical model provided by J. E. Littlewood could never do more than underpin the material evidence as assessed by Greg (whom Dane fails to remind us was a good mathematician), and that assessment—multiple editions of uniform size—proves unsustainable when account is taken of other factors such as absence of contemporary allusions, possible trial editions, aborted editions, proof-sheets, or the particular tastes of early collectors. In Chapter 2 Dane unpeels the assumptions about authorial revision arising in the accepted relationship between passages in version G (one manuscript) and version F (all other manuscripts) of The Legend of Good Women, demonstrating that their differences are likely no more than a matter of severe physical damage in the exemplar from which G was copied. Chapter 3 studies two cases in medieval vernacular drama, one where E. K. Chambers's rendering of the stage-plan for the Anglo-Norman La Seinte Resureccion proves to be the product of his theories about what the evidence should show rather than what it does show, the other what Dane terms the 'illusory history of production' (p. 53) that various editors have elaborated for Le Jeu du Pélerin. Chapter 4 is devoted to unmasking that 'authorial fiction' (p. 57), The Wakefield Master.

In Part 2, 'What is a Book?', Dane turns to early printed books and their later editors and collectors. The title of the key essay, Chapter 5, tells its own story: '"Ideal Copy" vs...