Manly and Rickert and the Failure of Method

Vance Ramsey, in his extended defense of Manly and Rickert's The Text of the Canterbury Tales, makes the following assertion: "Contrary to the absolutists, no intelligent person can doubt that learning, taste and judgment must be exercised by any editor worthy of the name, but they should be the very last resort of an editor and not the first."¹ I doubt whether anyone with experience of editing Middle English texts (as Ramsey had not) would formulate the problems of editorial intervention in such a categorical way. Nor is it clear what "absolutists" (whatever the term is intended to mean) he has in mind. And if "learning, taste and judgment" are to be the "last" resort of the editor, what is to be the first?²

It is worth applying this last question to the editorial activities of Manly and Rickert, that is, to those portions of their work specifically concerned with presenting a text of the Canterbury Tales, volumes 3 and 4. These volumes have important practical implications for their larger undertaking. One might reasonably assume that the very first procedural matter that any editor would wish to establish would be the choice of the base text, which will provide the lemmata for the edition, that is, the forms against which other witnesses will be collated. Normally such a base text would also provide, in general terms, the orthographic forms for the text, what are now termed its "accidentals."

The term "accidentals" derives from the influential editorial theory of W. W. Greg in his classic paper "The Rationale of Copy-Text."³ Although Greg draws his examples in this essay from Renaissance texts, his formulations drew on considerable experience of Middle English works and he had also made extended study of the problems of manuscripts of the Canterbury Tales.⁴ According to Ramsey, Greg had also been in correspondence with Manly and Rickert in the 1920s and 1930s.⁵ It is clear that the experience of medieval literature helped to shape his thinking.⁶ The idea of copy-text makes a clear distinction in textual authority between the accidentals (the spelling forms) of a base text and its substantive readings. Greg suggests that generally the earliest surviving witness (that closest in time to the author's original) provides authority for accidentals, while questions of substantive variation between witnesses can be assessed on a case-by-case basis using a full range of editorial resources (including doubtless "learning, taste and judgment").

Greg's formulation offers a clear and flexible statement of method: establish the choice of copy-text to provide a formal basis for a text, but emend its substantive readings on appropriate grounds. It is a statement of method that, in the case of the Canterbury Tales, invalidates the lengthy and fruitless debate over the textual superiority of Hengwrt or Ellesmere that Manly and Rickert conduct through assertion rather than the presentation of evidence.⁷ For any edition of the work must perforce take Ellesmere as its copy-text on wholly pragmatic grounds: it is early, dialectally consistent and, more compelling, it is the more complete of the two (Hengwrt lacks of course The Canon's Yeoman's Tale and much of The Parson's Tale and the Retraction), and it offers a more convincing order. Any debate about the superiority of particular readings in Hengwrt (or indeed other manuscripts) remains a matter of individual editorial judgment. Most editions now begin by identifying their "base text" or "copy-text" (the terms are often used synonymously). Such a statement might reasonably be seen as the "first resort" of any editor.

But not Manly and Rickert. Their textual methodology of course predates Greg's paper, but it has always been editorial convention to identify the text on which an edition is based. They do not. They do identify the grounds for one aspect of their textual procedure. They state: "As the basis for collation we chose Skeat's 'Student's Edition.'"⁸ Although they give no more details, the edition they used is presumably W. W. Skeat's one-volume Student's...