This essay focuses on New Bibliographical method and on its application to analysis of the particular texts - those of John Fletcher's play Bonduca - on which Sir Walter Wilson Greg depended for his influential conception of 'foul papers'. New Bibliographical method, in large part Greg's own creation, requires us to attend to what separates us as readers from our authors. As Greg once put it, New Bibliographical method concerns 'how a number of steps often intervene between the work as it formed itself in the author's mind and as it reaches modern readers'. For example, Greg went on, the New Bibliographer will 'describe the conditions under which manuscripts were . . . copied, the kinds of mistake that scribes habitually made, [and] the extent of corruption to be expected.' And Greg issued a dire warning to editors who fail to employ his method: 'Everywhere the editor suffers from not being a bibliographer; he gives himself all sorts of unnecessary trouble and arrives at all sorts of impossible results.'

My object is to demonstrate that in the essay where Greg developed his conception of 'foul papers', he himself did not employ New Bibliographical method, and as a consequence he arrived at impossible results. These have vitiated much editorial work on early modern drama during the period when his paradigm has been in force, despite formidable challenges to it by such comparable luminaries as R. B. McKerrow and Fredson Bowers. For most of that period it was not possible to identify Greg's crucial mistake because the essay in which he made it lay unpublished. Instead, in book after book including The Shakespeare First Folio (1955), Greg’s massive authority carried the day as he asserted his theory of ‘foul papers’. According to his theory, dramatists submitted their plays to theatrical companies as manuscripts too messy to be used to guide production, with the result that the companies had to have the plays copied in order to perform them. The companies, he argued, retained the authorial ‘foul papers’ and often provided them to the stationers who published plays. Cruces in these printed texts, on Greg’s theory, are to be resolved with reference to putative authorial sloppiness in the printer’s copy.