"Why Edit Anything at All?" Textual Editing and Postmodernism: A Review Essay

"Why Edit Anything at All?" Textual Editing and Postmodernism: A Review Essay Ian Small University of Birmingham George Bornstein and Ralph G. Williams, eds. Palimpsest: Editorial Theory in the Humanities Ann Arbor: University of Michigan Press, 1993. viii + 318 pp. $45.00 PALIMPSESTIS AVOLUME OFESSAYS by some of today's most distinguished textual critics. It had its origins in a conference held in the University of Michigan in 1991, and if the book is a fair reflection of that conference's proceedings, it must have been a particularly stimulating forty-eight hours. It is a consistently interesting collection which takes as its subject some of the key issues in literary studies today. Indeed, in keeping with a recent trend among editorial theorists, the volume claims that issues encountered by textual scholars are the central issues of the discipline, and in this sense it aspires to a larger readership than the ranks of textual-editors. Unfortunately, the book's sheer size and diversity makes a review which does justice to all its contributions impossible. However, one of the merits of the collection (and, by implication, of George Bornstein's and Ralph G. Williams's editing) is that we are invited to read it not just as a series of occasional pieces, but rather as an examination of "postmodernist " thinking about textuality and textual identity, issues which, as I have suggested, reach well beyond debates about the current state of editorial theory. Most of the individual essays are of an extremely high standard and well worth reading on their own account; my com195 ELT 38:2 1995 ments, however, will restrict themselves to what I perceive to be the larger argument of the volume. The title itself draws attention to the book's underlying thesis, that all texts possess a "multilayered" quality which is often hidden from the reader, but which can be investigated and illuminated by the modern textual scholar. In the words of George Bornstein's lucid introduction, "although we tend to think of major works as fixed or stable, a surprising number of them display upon examination a palimpsestic quality. . . . Increasingly, such works have come to seem contingent and constructed rather than unitary and fixed" (1-2). The issue at stake here is the opposition between postmodernist concepts of meaning—that is, its emphasis on indeterminacy, arbitrariness and artifactuality—»and those concepts of authority, intention and determinate meaning which have traditionally been used in literary studies. The editors suggest that textual editing must not only embrace postmodernism theoretically, but that it also must find ways of realizing its implications at a practical level. In this sense, Bornstein suggests, the metaphor of the palimpsest is useful, because it places an emphasis on what he terms "versions," rather than ideas of closure and stability (well represented, Bornstein claims, in the famous New Critical metaphor of the "well-wrought urn"). The text-editor seeking a cohabitation with postmodernism, though, encounters two kinds of difficulties. First there is the practical problem of how best to represent those "versions" in a neutral way in a scholarly edition, an issue which is aired in this volume by discussions of the potential of electronic texts; secondly, and I think more importantly, there are the ontological consequences of postmodernism itself. These last concerns are crystallized in questions of value and identity-<-in Peter L. Shillingsburg's words, the question "what then IS the work itself." Perhaps a simpler way of formulating all this is in terms of a series of issues which were once straightforward for traditional literary critics and editors. In what ways does a postmodernist epistemology problematize the familiar questions: "How do we edit texts," "Why do we edit the texts which we do edit," and "Why do we edit texts at alH* Few of the essays in this volume address all these questions, but read together they set up an interesting dialogue about them. Many of the contributors focus on the problem of authority and hierarchy; that is, whether the editor should represent the palimpsestic quality of a text—its "versions"—and, if so, how to achieve this ambition in a neutral or value-free way. Implicit in these questions are a set of political 196 SMALL : TEXTUAL EDITING judgments about the role and responsibility of editors vis-à -vis their readers: in simple terms, should editors aim to make value-judgements on behalf of'the reader; or should they rather aim to help readers make their own judgments? For D. C. Greetham, in common with several essayists, embracing a postmodernist epistemology entails the latter option: indeed, his use of terms such as "concealment" and "denial" suggests that editorial decisions have a moral dimension: [PJostmodernist editing operates under the assumptions of poststructuralist différance, the continued deferral of absolute meaning, and the texts it produces are scriptible not lisible, open not closed. External display of structure is not inevitable in such an enterprise, but it can be an advantage. . . . While it is true that externalization and display do not in themselves guarantee the reader's poststructuralist play in the text, and it is true as well that the converse (concealment of structure does not forbid readerly reconstructions ) is also not an automatic response to such texts, the almost universal adoption of concealment over display in eclectic editions cumulatively endorses a disjunct between text and apparatus inimical to the continued intertextual interpénétration of the two. The procedural denial of imported [misjquotation, from apparatus or elsewhere into the text is also a denial of the basic epistemology of textual criticism. (16-18) In the essay which follows, Peter L. Shillingsburg also demands that access to the work requires access to all its texts or embodiments or versions, and (like Greetham), he proposes this for the purpose of "freeing" the reader from the arbitrary authority of the editor; indeed Shillingsburg suggests that a rigorous pursuit of postmodernism should lead editors to "keep value laden terms like established, definitive, and total and complete out of their descriptions" (40). Moreover, both Greetham and Shillingsburg argue that the practical corollary of this political ambition exists in the power of the microchip: that—in Shillingsburg 's words—electronically-held text empowers readers to "identify the form, or forms, of the work they wish to deal with" (40). If this was all that was being said, the volume would not be at all noteworthy; after all, other textual theorists have been making similar arguments for some time. However Shillingsburg is also interested in the whole concept of what the "work" means—in the terms I used above, he is concerned with the relationship between text-editing, textual identity and ontology. In attempting to clarify that relationship, Shillingsburg offers, as a kind of thought-experiment, a postmodernist heresy (taken from George Steiner, an arch-idealist if ever there was one) that what we call the work "is not made up of any particular words 197 JSLT 38:2 1995 but, finding its expression in words of a variety of texts, the work is a spiritual presence that lives through earthen vessels and transcends the limitations of any particular embodiment of it" (38). Sbillingsburg's point is that what we recognize as the work seems to exist independently of any embodiments of it—that is, independently of any palimpsestic inscription. In his essay Shillingsburg offers this suggestion chiefly in order to demonstrate that a text existing in an electronic form does not present any more inherent disadvantages than one which exists in the more familiar codex form. This may indeed be the case, but Sbillingsburg 's "rhetorical questions" let Pandora out of the box. For if it is true that the work exists independently of any its embodiments (as a whole tradition of philosophical aesthetics, and most recently Richard Wollheim, maintain) then why do we need to edit anything at all? Moreover, if the "versions" do not impinge upon the identity of the work, then why does the reader need to know about them? In terms of postmodernist editing, Sbillingsburg's question—"What constitutes a work?"—amounts to asking, What constitutes the identity of a work, the identity of its versions and the nature of the relationship between them? Here the metaphor of the palimpsest turns out to be less useful than the editors suggest. The palimpsest recognizes not the work, only its versions; moreover it takes the identity of those versions for granted. So the "palimpsest" is simply there to be seen if we look hard enough for it. George Bornstein argues as much in his introduction to the volume, when he elaborates an opposition between versions, which are "there," and the notion of a final text, which is the (arbitrary) product of the values, choices and hierarchies of a certain kind of editor: A theory of versions tends to shift our conception of the artwork itself from product to process. Emphasis centers on the multiplicity of versions themselves rather than on privileging a final one to which the others seem mere stepping-stones. Seen in that way, the palimpsest becomes less a bearer of a fixed final inscription than a site of the process of inscription, in which acts of composition and transmission occur before our eyes. (3-4) This programme is reminiscent of Shillingsburg's desire to "free" the reader: postmodernist text-editing assumes that the presentation of "versions" is theoretically (but perhaps not practically) unproblematic and value-neutral. As Shillingsburg suggests, the reader's values will define the work. However, at this point there is a logical contradiction waiting to ambush the postmodernist editor. At its heart it concerns the 198 SMALL : TEXTUAL EDITING issue which Shillingsburg glimpses, but fails to resolve: the ontological relation between a work and its versions. The consequence of a postmodernist epistemology (as it is represented in this volume) is that "works" are constructed by the values and prejudices of the editor, but that "versions" are simply there to be revealed. In this view, then, there is no obvious relationship between a work and its versions. Rather the opposite: there is a systematic attempt to maintain that the two are quite separate, in the sense that it is assumed that the values which define the work are derived not from the versions, but from the editor or reader-as-editor. (It might be worthwhile noting in passing that "traditional" text-editing assumed a formative relationship between the two, in the sense that the editor's handling and priorizing of "versions" was solely in order to present—or in postmodernist terms, "construct"—the work.) If this suggestion is the case—that is, if there is no necessary relationship between versions and works—it is difficult to see why we need to know about versions, and more importantly, why we need to see all the versions simultaneously. Put more bluntly, if there is no logical connection between a work and versions, why bother with editions of any kind? But of course this postmodernist logic is not correct, and for one simple reason: the versions (that is, the palimpsest) are not simply "there" to be revealed. Versions, just like works, have to be identified, and that process of identification, like any other process of identification, willy-nilly involves values and prejudices. Thus our labelling of the world (and the literary- and art-worlds are no exception) can never be value-free. Indeed the recognition that the world does not come to us pre-labelled (and thus value-free) has been a staple of twentieth-century philosophy, and goes back at least to the work of Charles Peirce. The central question, then, is where do the values which allow the identification of the versions come from? The last essay in the volume, Clayborne Carson's account of the problems produced by the task of editing the work of Martin Luther King, Jr., suggests an answer. In this instance, the attempt to visualize the palimpsest revealed what Carson calls "intense and extensive" plagiarism in much of King's work. The point to notice, as Carson himself acknowledges, is that such a description is not neutral; the identification and labelling of the versions necessarily involve a simultaneous evaluation of them. It is not the case that textual features have been seen and at some later point labelled (or judged); seeing them and labelling them are coterminous 199 ELT 38:2 1995 activities, in the sense that what we find is always predetermined by a conceptual set. If a postmodernist epistemology is true, it is true of all knowledge: the identification of versions is as value-laden as the identification of works. Furthermore, the concepts of plagiarism and originality only make sense in relation to two further concepts, those of the integrity and identity of a work. In other words the identification of the versions (or, in the editors' terms, the apparently neutral description of the palimpsest) depends upon a prior value-judgment about what constitutes the work in question. To his credit, Carson makes no attempt to evade this dilemma and is open about the value-laden nature of his own editorial practice in identifying versions. The presence of other texts are described as "appropriations"—that is, as strategic, selft-conscious political borrowings: Recent literary criticism has made much of the ^determinant relationship between linguistic symbols and the reality they claim to symbolize. Likewise, King and other black leaders have long recognized that the democratic rhetoric they appropriate from white political leaders originally signified a racial reality African-Americans found unacceptable. The black leaders' achievement was in using the political vocabulary of the dominant culture creatively and ironically in order to change that reality. For King and other African-Americans, the appropriation of hegemonic texts was a political act. (313) Carson's practice articulates the postmodernist dilemma: for not only is a neutral presentation of "versions" impossible, it is also turns out to be undesirable. Indeed Carson feels that it his responsibility as an editor to make precisely those value-judgments which Shillingsburg explicitly exhorts editors to avoid. It might be thought that King's public reputation represents a special case, but what is true of King's works is true of all works. Works are important in a culture precisely because they reflect or embody one set of values and not another. If a work were multivalent in all its versions, it is hard to see how it could possess any cultural value at all, and why, therefore, any editor would want to edit it or, for that matter, any reader would want to read it. If individual readers were really free to make their own works, then literary art could have no social value or function because the whole notion of the social only makes sense in terms of collective activity (that is, shared, agreed and known values). In other words, we would have no need for editors or editions. If postmodernist editors pursued their principles with rigour, they would run the risk of putting themselves out of a job. But of course in practice, as Carson demonstrates, postmodernist editing 200 SMALL : TEXTUAL EDITING cannot be pursued rigorously, because the identification of versions— visualizing the palimpsest—is not, and can never be, a neutral activity. The argument which I am offering, then, is that versions are not simply "there" as the metaphor of the palimpsest suggests; rather, they have to be identified, and that process of identification depends upon a value-judgment about a work. Versions are always versions of something , and in this sense it is probably more useful to think about the term "versions" as simply that name which we give to a group of texts which are distinguished from other texts by virtue of the fact that we identify a relationship between them, and between them and a particular work. In this view, the concept of the work is logically coterminous with the identification of versions: that is, it is not possible to argue that versions and works can be separated for they are part of the same valuing activity. An example drawn from my own experience of editing Oscar Wilde's plays may make my argument clearer. The Herbert Beerbohm Tree Collection in the University of Bristol contains a wealth of documents relating to Tree's production of Wilde's A Woman of No Importance in 1893. One of those documents is what Russell Jackson and I have taken to be a working promptbook of the play which contains a list of the cast of the first production (in the Haymarket Theatre in 1893) and of a later revival. In other words, the promptbook relates both to the first run (and therefore may possibly have had Wilde's authority) and to a revival after Wilde's death. What claims do we make on behalf of this document when we allege that it is a version of A Woman of No Importance? Are the inscriptions within the promptbook (whatever they are) to count as a version? If so, what inscriptions exactly? Is the typed text a version in the same sense as the marginal comments and emendations are? Do all the emendations have the same status? Do emendations which refer to the revival matter in the same way as emendations which relate to the 1893 production? More pertinently, perhaps, how do we distinguish between these different sorts of emendations? In response, the postmodernist editor might argue these are exactly the sorts of decisions that should be left to the reader, that the editor's responsibility is only to reproduce the promptbook, a task which is now conceivable given the resources of hypertext. The difficulty with this answer is that it fails to acknowledge that the identification of the promptbook as a document worth reproducing is itself the result of a value-judgment. Russell Jackson and I initially selected the promptbook 201 ELT 38:2 1995 from all the other documents in the Tree Collection because we identified it as possessing a relationship with a work we knew as A Woman of No Importance. If we had not come to the library with an idea of what to look for—that is, those documents bearing a relationship to a particular work—we literally would not have known what to do. The only reason I know that an article in a nineteenth-century periodical is not a version of a particular Wilde play is because I carry in my head a concept of the work (the play) which allows me to generate a closed set of the features which define it. Of course it is possible that my concept of the work will change over time because there is a dynamic relationship between the work and versions. But the important point to bear in mind is that there always are (and there always must be) boundaries to that relationship. This circumstance in turn is simply a consequence of human cognition, of the ways in which we label the world. The dynamic nature of this relationship can be illustrated by another example. In 1991 I was in the company of Peter Raby in the William Andrews Clark Memorial Library when he discovered a document which he identified as the earliest known scenario of The Importance of Being Earnest.1 Most of the details—of character, plot, and so forth—contained in the scenario were quite different from the work which we recognize as The Importance of Being Earnest. So how did Raby know that this document was indeed a version of Wilde's most famous play? In practice, the judgment was immediate and intuitive, and one which all present concurred in. But if we tried to dismantle that agreement, we would find a very complex process of evaluation at work in which the scenario was compared not simply to a printed text of The Importance of Being Earnest, but to the whole panoply of versions which had already been identified. Some of those versions resemble the scenario as much as they do the printed text. Perhaps the way to understand this cognitive process is not via the metaphor of the palimpsest, but by means of Wittgenstein's notion of family resemblance. At the very least, we can say that the identification of version is not a rule-governed activity. Like critical judgments, it depends upon a complex set of evaluative processes . The point to emphasize in all this is that the postmodernist editor works under exactly the same conditions: without a concept of a work to guide the identification and selection of versions, such an editor has no choice but to reproduce everything written, literally every mark on every page. Indeed in the case of a play, which is a social text, why not 202 SMALL : TEXTUAL EDITING also include every performance? Such a Herculean (and ultimately pointless) task ought to make even the most liberal-minded editors stop in their tracks. My example is admittedly an extreme one, but it has the advantage of illustrating very starkly a truth which all editors, postmodern and traditional, must face up to: we simply cannot escape value-judgments. In this sense it is significant that in other areas of the discipline some critics, interested in the origins of art, have begun to turn away from the postmodernist project in order to revive older concepts of aesthetic integrity and aesthetic identity, because it is only such concepts which permit an understanding of the social functions and value of cultural artifacts: that is to say, a recognition of the social nature of art and literature logically requires a recognition not of arbitrariness and indeterminacy, but of precisely their opposites.2 Art and literature cannot have social functions unless, at some point in time and for some readers, they possess determinate meanings—that is, they exist as works, and not as versions. In this respect it may be that the real responsibility of the editor is not to give an illusory freedom to the reader, but rather, as Claybome Carson does, consciously and openly to make value-judgments on behalf of the reader. Perhaps the socially useful task of the editor is to continue to do what generations of editors have been doing—not attempting to make "final" texts, but producing works useful for their readers in their time. Notes 1. See Peter Raby, The Making of The Importance of Being Earnest," Times Literary Supplement, no. 4629 (December 1991), 13. 2. A full discussion of this development is beyond the scope of a review essay. However, my own interest in this line of argument, developed in collaboration with a colleague, may be found in Josephine Guy and Ian Small, Politics and Value in English Studies (Cambridge: Cambridge University Press, 1993); other recent works which I judge to be of a similar spirit include: Claire Badaracco, The Editor and the Question of Value: Proposal," TEXT, 1 (1984), 41-43; Stein Haugom Olsen, The End of Literary Theory (Cambridge: Cambridge University Press, 1987); Colin Falck, Myth, Truth and Literature (Cambridge: Cambridge University Press, 1989); and David Novitz, The Integrity of Aesthetics,* Journal of Aesthetics and Art Criticism, 48 (1990), 9-20. 203 ...