 | |  |
Textual Instability and Editorial Idealism
by
G. Thomas Tanselle
Over the past decade and a half there has been a notable shift of emphasis in the writing about textual matters, from a concern with authorial intention to an interest in the collaborative or social aspects of text-production. As the discussion has evolved, there has come to be an increased concern with textual instability and the significance of versions. All these concepts are naturally linked: for if texts are social products, then texts will take different shapes as they pass from one social milieu to another; and if authors are not the only source of validity in the constitution of texts, then all these variant texts carry their own authority as products of history. This attention to textual multiplicity extends backward into the initial creative process and forward into audience response (which is also a creative activity). Authorial versions — obvious emblems of textual instability — are studied both as products of particular moments and as parts of an endless process, which does not cease with the author's death; and readers, both during an author's lifetime and afterward, participate in this process by creating their own versions of the texts they encounter.
These ideas are obviously related to the anti-foundationalist tendencies of philosophy and literary theory in the past generation. Their presence in recent textual theory is to be applauded as a sign — too often lacking in the past — that textual and literary theory and criticism share the same concerns; it also serves to focus textual critics' attention on topics they had previously neglected. There is no reason, of course, why these new interests should involve a banishment of authorial intention as a subject of study: it is clearly a part of the total textual process, and any comprehensive approach would have to include it. Many of the recent discussions, however, have denigrated it and have presented a distorted picture of intentionalist editors' attitudes. A number of stock elements reappear in essay after essay: intentionalist editors in the "Greg-Bowers tradition," so the standard argument goes, unrealistically think of a literary work as the isolated product of a single individual, and they seek to replace textual instability with fixity by conflating variant texts
In the pages that follow I shall survey the general theoretical writings that have appeared in English during the first half of the 1990s, trying to sort out what remains valuable after the distortions, oversimplifications, and loose arguments are removed. Because the field of textual study is attracting more interest as its connections with literary theory are more widely noticed, the body of writing from this five-year period is extraordinarily large. I have necessarily been selective, omitting, for example, most of the articles dealing with particular authors or editions.[2] Introductions to textual study are a similar category in that they do not normally aim to break theoretical ground, but I should note in passing that both D. C. Greetham and I produced new introductions during the past several years. Greetham's Textual Scholarship: An Introduction (1992; reprinted with an expanded checklist in 1994) takes the opposite approach, quantitatively, from my A Rationale of Textual Criticism, published three years earlier. His book is a long one, attempting
An event during these years that will not have gone unnoticed by anyone connected with textual scholarship was the death of Fredson Bowers on 11 April 1991. His dominance of the field for much of the second half of the twentieth century naturally made his death seem symbolic, coming at a time when the kind of edition he championed was under attack.[7] His view of the writings surveyed here might not have been very favorable, but the reason would not have been their frequent criticisms of intentionalist editing, for he was always open — as his editorship of Studies in Bibliography showed — to points of view that were not his own. What would probably have bothered him is the emphasis on theory, since his own preference was to concentrate on the practical problems of dealing with particular situations.[8] Theory is, I believe,
This is one of the phenomena illustrated by the writings discussed below. I shall first examine two books, one by Jack Stillinger and the other by Jerome J. McGann, that exemplify the kinds of arguments being made and provide the opportunity for demonstrating the kinds of responses that seem to me appropriate. Dealing with these books in some detail will establish the framework within which one can respond to other writings and will, I trust, make it appropriate to comment more briefly on the other main contributors to the current debate. The second section below takes up the anthologies that were a notable feature of the early 1990s, and the third discusses some of the other essays, not in anthologies, that try to clarify the meanings of, and the distinctions among, concepts like work, text, and version. (Such discussions form a genre that looms large in the recent literature.) The final section tries to assess where all this commentary leaves us and how we can proceed.
I
The title that Jack Stillinger chose for his 1991 book — Multiple Authorship and the Myth of Solitary Genius — reflects a strain of thinking that is one of the most characteristic features of recent textual criticism and reveals Stillinger's polemic stance. The first half of the title refers in a straightforward way to the collaborative aspects of all verbal works (indeed, all human creations); the second suggests that the discussion of this topic is meant to expose a common misconception, a "myth," which is then expressed in slanted terms as a belief in "solitary genius." Whether the idea of single authorship (to use a more neutral phrase) is a myth depends of course on how it is defined. Certainly it can be defined in such a way that it is not a myth, and I would argue that its commonly understood sense, among general readers and textual scholars alike, is far more realistic and sophisticated than the label "solitary genius" implies. Stillinger has not avoided the argumentative trap of treating the object of his attack in a reductive way. The bulk of his book presents a number of case histories of multiple authorship, and the work
Signs of Stillinger's reductivism are evident in the opening chapter, "What Is an Author?" He recounts the textual history of Keats's "Sonnet to Sleep" in order to demonstrate that "actually Keats wrote only most of the words — not all of them — and in the course of revision, transcription, and publication, the sonnet underwent numerous changes" at the hands of several other persons (p. 17), whom he also denominates "authors." Thus the text of the sonnet in Richard Monckton Milnes's Life of 1848 "has at least three authors" (p. 19), and the version in H. W. Garrod's Oxford edition "has at least four authors" (p. 20). Indeed, Stillinger adds, "I am myself part-author of the text of the sonnet," because in his 1978 edition he made some further alterations to the text. The subject of his book can therefore be stated as "situations where someone other than the nominal author is essentially and inextricably a part of the authorship" (p. 20); and he asserts, as if it were a startling thesis, "I wish to claim that such multiple authorship — the collaborative authorship of writings that we routinely consider the work of a single author — is quite common" (p. 22).
There is no question that one can define authorship in this way if one wishes, using "author" to refer to anyone who plays any role at any time in the constitution of a text. But virtually no one would be surprised to learn that, under this definition, all texts have a multiple authorship. The concept of "single author" that most people "routinely" employ is not at odds with this point, for scarcely anyone is so simple-minded as to believe that a work by a single author does not incorporate any element that originated outside that author. What should bother any thoughtful reader of this chapter is that the most challenging and ultimately significant aspects of defining authorship are bypassed. To contrast multiple authorship, in Stillinger's inclusive sense, with the theoretical opposite extreme, a totally uninfluenced person, is not very fruitful. A far more intriguing pursuit would be to try to distinguish the kinds of collaborative influence that are realistically integral to the idea of an individual author from those that are at odds with it. Stillinger includes in his enumeration of "standard modes of composite creativity" not only the influence of friends but also "the author revising earlier versions of himself" and "the author interacting collaboratively
Instead of offering a lucid framework for thinking about the varieties of collaborative action, this introductory chapter is bound to confuse readers. After ostensibly suggesting that each version of a text should be accepted on its own historical terms (with its authorship cumulatively incorporating all who have affected its makeup), Stillinger at another point seems to imply that authenticity lies with the originating single author: when Keats approved of a change made by Woodhouse, the new reading, we are told, "becomes an integral, rather than a corrupt, element of the text" (p. 21). But even without Keats's approval, it would have been an authentic (or "integral") part of the Woodhouse version, in which one can legitimately be interested. What is missing in the discussion is any clear outline of the range of interests that historically minded readers can have. One might wish to focus only on the originating author's intention (which cannot realistically exclude influences willingly and actively incorporated into it by that author); or one might choose to examine the form or forms of the text that emerged from the process of publication (forms that might include features not sanctioned, or even actively disapproved, by the originating author); or one might study any posthumous version as a way of understanding the responses of its readers. The second and third of these approaches concentrate, almost inevitably, on collaboratively produced texts that have survived in documents; the first necessitates the effort to reconstruct what the originating author intended, even if those intentions have been subverted by the documentary texts. Such a reconstruction, as traditionally undertaken by intentionalist editors, involves the interesting question of how to judge the spirit in which an author responded to various external stimuli. As in all matters of judgment, no single answer will satisfy everyone; but the range of serious answers reflects the struggle to locate the boundaries of individual creative activity. An authorially intended text, as normally conceived, includes readings that result from willing collaboration and excludes those produced under duress.
Stillinger's gross oversimplification of these issues is clear in his claim that the ubiquity of multiple authorship is "rather strikingly at odds with the interpretive and editorial theorists' almost universal concern
In his final chapter ("Implications for Theory"),[11] he returns to a consideration of editorial theory and devotes the closing pages of the book to a criticism of the "Greg-Bowers school" (p. 200). Whereas readers unacquainted with editorial debates of the past half-century will nevertheless see the flaws of the opening chapter, they will not be in a position to recognize the distortions in his presentation of the Greg-Bowers approach in the last chapter. He repeats the hackneyed claims that this approach rigidly applies the same rule to all situations, that it always combines early accidentals with late substantives, and that it is inhospitable to the idea of independent versions (see pp. 196 — 198) — claims that cannot be substantiated.[12] But even those readers (untutored
The key to the problem surfaces in an earlier comment: "authors' intentions are no more available to editors than they are to interpreters" (p. 195).[13] If collaborative intentions are considered to be "available" because the texts that have come down to us in physical objects are likely to be collaborative, and if only such "available" texts are proper subjects for historical study, then two fallacies of historical methodology are implicit here. One is that artifacts speak clearly for themselves; the other is that surviving materials are necessarily more instructive than speculations about those now lost. As to the first, a documentary text does not make the collaborative intention behind it "available" without analysis and interpretation (since the problem of distinguishing intended from unintended aspects of physical texts is equally acute no matter how many "authors" there are). As to the second, other texts, besides those now surviving in physical form, once existed (on paper, perhaps, and in people's minds, certainly), and some of them may be of enough interest to be worth attempting to reconstruct; the process requires judgment, but no more so (and no less so) than the use of documentary texts.[14]
To say that intentionalist editors produce "texts that never previously existed" is not a criticism (if it means, as I assume it does, "never previously existed on paper"), for there is no reason to suppose that intended texts have always appeared in physical form — nor is there any reason to banish from historical inquiry the use of informed judgment to reconstruct past events. The claim that reconstructed texts realize "editors' rather than authors' intentions" would, if taken literally, be a valid objection (so long, that is, as the concern is with historical reconstruction rather than the unending process of textual collaboration). But what such texts normally incorporate is editors' judgments about past intentions rather than editors' own intentions. Editors' own temperaments are naturally involved, but only in the same way that all historians' personalities play a role in the attempted recreation of the past. Stillinger seems, surprisingly, to have adopted here an extremely restrictive concept of what constitutes proper historical scholarship.[15] He was of course under no obligation to deal with these matters, and it is unfortunate that he chose to do so. His studies of specific instances of "multiple authorship" are informative and valuable in their own right; by deciding to frame them within an attack on intentionalist editing, and then conducting the argument illogically, he weakened his book.[16] But the urge to narrow the range of legitimate editorial goals was obviously hard to resist at the time he wrote.
The name of Jerome J. McGann comes up in Stillinger's book as the exponent of a "socialized concept of literary production," with which "the facts of multiple authorship are most compatible" (p. 199). A collection of McGann's essays, entitled The Textual Condition, appeared the same year as Stillinger's book (1991) and offers a far more effective statement of the position. Although McGann claims that his book provides "case studies" and not "a theory" (p. 16), readers are likely to feel
All this is unexceptionable in itself. Where McGann falters, as his earlier writings have already demonstrated, is in his linking of these ideas to a rejection of authorial intention as a goal of textual study. In this instance, he claims that Paul De Man and I "come together as textual idealists," for we both (despite other pronounced differences) are caught — as he sees it — between a concept of artistic creation as "a transcendence of the human" and a realization "that the terms of such a demand can never be met" (p. 7). He contrasts his own approach as "another way of thinking about texts" (p. 8) — an alternative that he apparently sees as incompatible. The exposition is confrontational: "One breaks the spell of romantic hermeneutics by socializing the study of texts" (p. 12). Any point of view that holds a spell over a field no doubt deserves to be challenged, and questioning authorial intention as the dominant concern of textual criticism has been a salutary phenomenon. But the questioning is less helpful than it might be if it fails to recognize how the intentionalist and the socialized emphases fit together — if, in other words, it simply replaces one spell with another, for to do so continues to restrict artificially the areas of investigation that can profitably be pursued. I think one can see how McGann distorts the study of authorial intention if one looks at two matters: his associating it with "textual idealism," and his seeming rejection of intended texts as stages in the history of textual instability.
As to the first, McGann confuses "idealism" with a notion of perfection as well as with a concern for "idea" rather than execution. His "other way" of looking at texts, he explains, rejects any sense that texts are less perfect as a result of being embodied physically; embodied texts are examples of "perfect limitation," their "perfectness" being "located within the sociological particulars which the perfection defines" (p. 9).[18] To believe otherwise — to hold that texts in physical form are debased versions of "ideal" texts — is to believe, he says (quoting Shelley's famous words), that "the most glorious poetry . . . communicated to the world is probably a feeble shadow of the original conception of the poet" (p. 8). Or, as McGann argues at the end of the book, if "text" has "no material existence" and "comprises an abstraction," then "text" is being taken to mean "the idea of textuality" (p. 178, italics his).
I do not believe, however, that any of these points has been held by the editors who have concentrated on authorial intention: their goal is not to reconstruct the "idea" that lies behind a work but to recover an actual text — a specific set of words — that is not adequately represented in any known physical document. My phrase "the intractability of the physical," which McGann takes as a sign of my "romantic" position (pp. 7 — 8), has — in its context in A Rationale of Textual Criticism — a less exalted meaning. It simply refers to the difficulty of getting words transferred accurately to a physical surface. Authors do formulate texts, not just ideas for texts, at the moment of composition; but they may make mistakes in writing down the words. An authorially intended text is a text that once existed, though it may not have existed in physical form. Such a situation can occur because language is intangible, and a verbal text can therefore exist apart from being made physical. Although literary works may frequently fail to live up to their authors' ideas for the works, this point is irrelevant to what intentionalist editors do, for they are concerned with the works, whether or not those works are pale shadows of grand conceptions. The only sense in which intentionalist editors construct "ideal" texts is that those texts may not have existed in physical form before the editors produced them; but such editors do not think of their texts as "perfect" in any sense, nor do these editors believe that they are uncovering the "idea of a text" underlying any particular executed text. Intentionalist editors are not idealists in a philosophical sense (or, at least, are not revealing in their work where they stand on this matter), because their activities do not imply a belief that truth lies
We thus come to the second of McGann's misconceptions about the study of intentions: his failure to consider intentions as historical events. The history of a text begins before the social encounter in which a reader engages a pre-existing text (pre-existing only in the sense that the reader is interacting with something and not in any sense that attributes passivity to reading). One can, if one chooses, define the condition of textuality so that it is limited to collaborative interchange, but doing so excludes part of the story. Surely the "textual condition" must include what authors face when they are working at their desks; and the texts they form in their minds have a place in any encompassing view of textual metamorphosis. The instability of texts is beyond dispute, and editors who produce critical texts aimed at representing authors' intentions have never claimed otherwise — as their repeated comments on the importance of an apparatus of variants shows.[20] They construct such texts not because they believe that no other texts are valid but because intended texts are not available in documents and therefore cannot be studied without the exercise of critical judgment, leading to attempted reconstructions. The most fundamental point underlying this activity is a recognition that historical inquiry cannot be limited to what happens to exist in physical form. Any kind of history — not merely textual but also military, economic, political, social, and so on — depends on the effort to fill the gaps in, and indeed to correct, the physical record, which is inevitably incomplete and potentially misleading. Intentionalist editors are not, through their work, declaring that an ideal (and unchanging)
In between the "Introduction" and the "Conclusion," on which these comments are based, there are seven essays (five of them previously published, some in shorter form) that similarly illustrate the limitations of McGann's argument.[22] His considerable strengths are shown, too, largely
The difference between the two approaches can be further clarified by using the terms of Chapter 2, "What Is Critical Editing?", where one of the main defects of Greg-Bowers critical editing (in his view) is defined as its failure to deal with the "bibliographical" (or physical) characteristics of textual artifacts. "The weakness of the theory," he says, referring to copy-text theory, "is that it largely ignores the transmissive or communicative aspects of linguistic events" — for "'Copy-text,' in modern editorial theory, is always a linguistic text" (p. 57). By regarding textuality as the condition emerging from an intertwining of linguistic and bibliographical "codes,"[23] and by recommending an emphasis on the comparative analysis of variants rather than on the production of eclectic texts, McGann believes that he is offering "a more comprehensive imagination of the fields of textual criticism and critical editing" (p. 50, italics mine). Just above I objected that his approach lacks the "comprehensiveness" of the one he is criticizing. How these cross-claims of comprehensiveness arose is not difficult to explain and defines the central issue of the debate. If critical editing in the Greg-Bowers tradition — or "eclectic" editing, to use McGann's term — were really dismissive of, or oblivious to, the textual significance of physical presentation,[24]
A brief comment should perhaps be added here about the first chapter, entitled "Theory, Literary Pragmatics, and the Editorial Horizon," because it shows how readily the intentionalist approach actually fits within McGann's overall scheme. After making his usual point that "texts are produced and reproduced under specific social and institutional conditions," he concludes that "every text, including those that may appear to be purely private, is a social text" — an "event" or "point in time" (p. 21). Those who focus on authorial intention (and are thus interested in texts that may, at first thought, "appear to be purely private") are in fact attempting to reconstruct particular events that occurred at particular times, events that are inevitably social, since intention is itself socially formed (as I pointed out in discussing Stillinger above). But instead of giving intentionalist editing its niche within his
An editor who emends an early documentary text with some readings from a late document is not attempting to construct a single, final, ideal, ahistorical text but to reconstruct what was intended at a specific time — either an early time (if the argument is that those emendations represent what had been intended but not executed in the early text) or a late time (if the argument is that, with the exception of the late readings chosen as emendations, the text of the late document does not reflect the author's late intention as satisfactorily as does the text of the early document). And the process, though perhaps "technical" and "specialized" in some respects, is certainly not mechanical, since it involves critical judgment throughout. The result is therefore undeniably a product of the present, but it is a present effort to recapture the past, which is all that any historical scholarship can be, as McGann well understands. When he says at the end, "To edit a text is to be situated in a historical relation to the work's transmissions, but it is also to be placed in an immediate relation to contemporary cultural and conceptual goals" (p. 47), one can scarcely disagree with him. And I am glad that he labels as "deeply mistaken" the idea of the "editor-as-technical-functionary." But it is regrettable (and puzzling) that he encumbers his basically sensible point of view with an unreasoned exclusion of one whole area of textual history, thereby turning his book into a symbol of the recent fixation on intentionalist editing as supposedly based on unrealistic notions of literary production and textual stability.
II
Perhaps the most distinctive feature of the publishing scene in the field of textual criticism during the early 1990s was the appearance of an extraordinary number of anthologies of essays — so many that this period may be thought of in the future as the Age of Anthologies in textual study. From 1990 through mid-1995, at least twenty-five anthologies in English (or largely in English) were published, in addition to the three volumes of Text (volumes 5, 6, and 7) brought out by the Society for Textual Scholarship during this time.[28] This figure can be put in perspective by noting that only about thirty-four were produced in the entire two decades preceding 1990 (plus four volumes of Text). Of those, nineteen were part of the series of annual volumes (in effect constituting a periodical) containing the papers delivered at the Toronto Conferences on Editorial Problems; since only three Toronto volumes appeared in the early 1990s, the dramatic nature of the recent increase in anthology production can more truly be shown by comparing the numbers of anthologies excluding the Toronto series (and volumes of Text) — some fifteen during 1970 — 89 and twenty-two during 1990 — 95.[29] For many years — through 1980 — there were essentially only three separately published anthologies (besides the Toronto volumes, which began in 1966), and they consisted almost entirely of previously published pieces;[30] the gathering of original essays did not begin in earnest until the 1980s, when several notable collections appeared.[31] But the real spurt occurred in the early 1990s and (since the number of periodicals available as outlets had
These recent anthologies have published a large amount of valuable textual commentary, but fewer than half contain the kind of general theoretical discussion that requires examination here. It is not necessary, for example, to comment on those collections that concentrate on a single author or edition, such as the New Oxford Shakespeare or the Gabler Ulysses.[33] Then there are those anthologies that deserve a mention in this context because of their significant treatment of bibliographical study at large but that are not primarily concerned with textual work — namely, Peter Davison's The Book Encompassed: Studies in Twentieth-Century Bibliography (1992) and Nicolas Barker's A Potencie of Life: Books in Society (1993).[34] Some of the anthologies dealing with a single period or country can also be passed by, though it is worth noting that they frequently allude to the recent theoretical debates and the growing interaction between literary and textual criticism. For example, the introduction to the collection of papers from a 1989 Odense symposium, significantly entitled The Medieval Text: Editors and Critics (1990), speaks in its second paragraph of "the intimately intellectual relationship between medieval as well as modern editors and critics" (p. 9).[35] In another volume with a similar title, Tim William Machan's Medieval
Of the 1990 anthologies, two — both based on conferences held in the spring of 1989 — include general theoretical papers by prominent commentators. New Directions in Textual Studies, edited by Dave Oliphant and Robin Bradford, results from a conference of the same title at the University of Texas in Austin (30 — 31 March, 1 April 1989) and deals — in the words of Larry Carver's introduction — with "the fate of the Greg-Bowers model" and "the place of the social setting in the editing of books and manuscripts" (p. 10).[40] Articles by Jerome McGann, Randall McLeod,
The other anthology, Editing in Australia, edited by Paul Eggert, emerges from an April 1989 conference at the Australian Defence Force
The other essay in Editing in Australia that requires comment here[44]
One is therefore surprised by his apparent belief that there is a choice to be made between "product" and "process," as when he declares himself in favor of the latter (p. 46) and notes approvingly that "editions which emphasise the importance of Process" are "gaining ground" (p. 62). The two concepts are complementary, for every "product" is one element in a "process": to think in terms of process entails thinking of a series of products. And every conceptualization of a text, as envisioned by Shillingsburg, results in a product. Critical editors in the past often "foregrounded" one such product as a result of the limitations of the codex form of their editions, but they generally intended their apparatus to situate that product in a process, any stage of which was worthy of attention.[45] Of course, some editors have undoubtedly made "extravagent claims for the correctness of the Product" (p. 62), and their attitude can properly be objected to; but one cannot assume that the presence in an edition of a "single-finished-product text" (p. 45) points to an editor who fails to understand the importance of studying textual process.[46] Nevertheless, Shillingsburg's essay contains more admirable than
The Shillingsburg and Eggert essays were reprinted the following year in Philip Cohen's anthology Devils and Angels: Textual Editing and Literary Theory, which aims to promote dialogue between editors and literary theorists by exploring the ways that recent developments in literary theory have "problematized" (p. x) traditional editing and set in motion what may be "a paradigm shift in textual criticism" (p. xiv).[47] The essays are organized into three groups, each ending with an article of response to the essays in that group. In the first such response, T. H. Howard-Hill displays a useful skepticism regarding the essays by Shillingsburg and McGann (the McGann piece discussed above as the first chapter of The Textual Condition) and offers some needed reminders that the basic points currently being debated are not "fresh issues" (p. 48) — indeed, he says, Greg recognized the instability of texts "before McGann, Shillingsburg, and I were born" (p. 49).[48] Steven Mailloux, in the second response-article, discusses the Eggert piece, an essay by D. C. Greetham, and one written jointly by Philip Cohen and David H. Jackson.[49]
As for the essays thus commented upon, the most important (and best-written), in my view, is McLaverty's — indeed, I would place it among the four or five best essays on textual theory in recent years. He suggests thinking about textual instability in terms of two questions, which indeed go to the heart of the matter: what are "versions," and what texts should editors present to readers? He defines a version as an "utterance" (the "social act" [p. 144] of "making outer or external" [p. 140]) that is linked to one or more other utterances through "material, structure, and function" (p. 137). The history of a work, extending through time, is likely to comprise numerous utterances, all of which are obviously relevant to understanding that history. But, as McLaverty perceptively argues, the history of the work is to be distinguished from the text of the work, which is not the sum of all the utterances but the text of every utterance individually considered. What an edition should present, he recommends, is one or more of "the author's intended utterance(s)," with an apparatus enabling the reader "to construct the relation of each
The essay of Grigely's has many links with McLaverty's, but Grigely — drawing even more widely on recent literary theory and philosophy — emphasizes the unrepeatability of utterances, since every utterance is an event; even if the words can sometimes be accurately repeated, the context cannot. With wit and intelligence (and a good dose of jargon), he sketches here what he hopes is "the beginning of a philosophy of textuality" (p. 192); but it also may seem to imply an end of editing. Yet editors, like other historians, have always known that they cannot fully recapture the past; and Grigley's comprehensive concept of utterances can in fact help clarify for them what their efforts do, and will continue to, accomplish.[52] Of the other essays, Greetham's energetically explores "the ideology embedded in form and method" (p. 81),[53] Gabler's scolds
The other anthologies of 1991 contain fewer general theoretical essays. Representing Modernist Texts: Editing as Interpretation, edited by George Bornstein, largely consists of pieces on the texts of "modernist writers" (Yeats, Pound, Joyce, and Woolf, among others); but it does contain an introduction by Bornstein entitled "Why Editing Matters" (pp. 1 — 16) and a concluding essay by Michael Groden entitled "Contemporary Textual and Literary Theory" (pp. 259 — 286). Bornstein states that the volume aims "to explore the implications for literary critics and theorists of the recent revolution in editorial theory" (p. 5); and though he sometimes exaggerates the contrast between "traditional" theories (which "postulate the notion of an ideal 'correct' text" [p. 5]) and newer approaches (which "dislodge the notion of one privileged form for a text exercising authority over all other forms" [p. 7]), he usefully emphasizes "a firm belief in the potential of contemporary literary criticism and theory and of contemporary textual scholarship and theory to enrich each other" (p. 9). A volume exemplifying this belief is certainly to be welcomed, as is Groden's contribution, which perceptively and thoughtfully examines recent textual theories. In contrast to many of the historical surveys that now exist, his does not oversimplify the positions discussed, and it offers the best account I have seen of the relations between intentionalist editing and the New Criticism.[55]
Another 1991 collection that focuses on the literature of a single period is the special issue of Romance Philology on medieval textual criticism (45.1, August 1991), edited by Charles B. Faulhaber and Jerry R. Craddock (and containing one of the most extensive checklists of writings on textual criticism and on the use of computers in textual criticism [pp. 206 — 236]). Of the overviews of the French, Italian, and Spanish fields, only the one on Old French is in English, and it is distinguished: Mary B. Speer's "Editing Old French Texts in the Eighties: Theory and Practice" (pp. 7 — 43). Speer lucidly and sensibly summarizes the twentieth-century history of the field (following from Bédier's advocacy of the best-text approach), offers a shrewd analysis (pp. 15 — 22) of Bernard Cerquiglini's Eloge de la variante (1989), and examines a number of recent editions, recognizing that editions always reflect "an interpretive conception" (p. 25). She finds the "intensity" of recent debates "a healthy sign" (p. 24), and her balanced way of dealing with them merits the attention of scholars outside, as well as within, her field.
Editorial subjectivity again receives considerable attention in Ian Small and Marcus Walsh's collection, The Theory and Practice of Text-Editing: Essays in Honour of James T. Boulton (1991). The introduction, by both editors, and the concluding two essays — Walsh's on Richard Bentley and Small's on annotation — deal with some of the ways that judgment permeates the editorial enterprise and effectively illustrate how any aspect of editing that one examines can lead to a consideration of basic questions about the nature of literature.[56] Annotation is the entire subject of another 1991 anthology (based on an April 1988 Irvine conference), Annotation and Its Texts, edited by Stephen A. Barney. Although the essays cover all forms of scholarly commentary (or "secondary discourse," to use the term in Jacques Derrida's concluding essay), not just the apparatus for editions, the relevance of the topic to the study of editorial subjectivity is obvious. Barney's brief introduction repeatedly speaks of "a politics of annotation" and "the rhetoric of annotation," reflecting the same interest as recent writers on textual criticism, who
In 1993, George Bornstein and Ralph G. Williams edited Palimpsest: Editorial Theory in the Humanities (based on a November 1991 Ann Arbor conference) as the first volume of a series called "Editorial Theory and Literary Criticism" to be published by the University of Michigan Press (publisher of Bornstein's 1991 anthology). Bornstein explains the choice of the layering metaphor of the palimpsest for the title as a way of emphasizing the recent focus on works as "contingent and constructed rather than unitary and received" (p. 2). In order to link the Greg-Bowers "consensus" inflexibly with the latter, however, he grossly exaggerates its "absolutist claims." The concept of authority, he believes, is basic to the distinction, for authority "seems to recognize a stable, unitary text rather than an unstable, multiple one." But his examples of textual authority involve the perceived need on the part of religious, political, and social institutions to promulgate "authorized" texts or to "legitimate" themselves by so doing;[57] and in limiting "authority" to this sense, he ignores its primary meanings for textual scholars. Documentary texts are usually said to possess authority when they provide relevant evidence for a given purpose: thus for editors concerned with authorial intention, any documentary text that can serve as the most direct source of an authorial reading is a text carrying authority. There is no suggestion here of a "unitary" text of authority (because many texts with authority may exist), and there is no reason for "authority" in this sense to be of use only to intentionalist editors. Furthermore, when critical editors apply "authority" or "authoritative" to their own constructed texts, they do not usually mean that those texts are the only ones required, for every purpose now and in the future; what they generally mean is that the critical texts result from the systematic effort to identify and evaluate
This introduction sets the tone for much of the volume. In the opening essay, for example, D. C. Greetham attempts, with characteristic playfulness, to distinguish editorial modernism from postmodernism — but at the price of distorting "twentieth-century eclecticism" by equating it with an "attempted reconstruction of the 'text that never was,'" or with a "courting of the ideal authorial form" (p. 18). Eclectic editors, however, explicitly seek to construct what once existed and do not think of the result as "ideal" (whatever that means) — nor do they believe that their efforts "will produce a final 'understanding' of the text." Greetham's contrast — between the modernist editor striving for a stable, authoritative text and the postmodernist editor emphasizing the open-endedness and contingency of texts — is unrealistically neat and compartmentalized. In the next essay Peter Shillingsburg makes generally helpful observations about electronic texts, but not without first claiming — incorrectly — that scholarly editors have traditionally believed a literary work of art to be "equivalent" both to "the linguistic text" and to "the ideal or best version of it" (p. 31).[58] But Palimpsest does contain a counter-balance in the form of Ralph G. Williams's reasonable and unpolemical meditation: in concluding that textual boundaries and meanings are indeterminate and that no philosophical basis exists for believing in the identifiability of intentions, he does not oversimplify the positions held by intentionalist editors. Indeed, he recognizes that the inability to determine intentions with certainty does not make the search "uninteresting, illegitimate, and passé": "We have protocols of considerable subtlety for evaluating intent, and that game seems not only a persistent
Another anthology of 1993, W. Speed Hill's New Ways of Looking at Old Texts: Papers of the Renaissance English Text Society, 1985 — 1991, collects — as the subtitle indicates — the papers that were read at seven annual meetings of RETS and prefaces them with a substantial essay of Hill's entitled "Editing Nondramatic Texts of the English Renaissance: A Field Guide with Illustrations" (pp. 1 — 24).[60] The essay is an extremely useful survey of the field despite its bias against the Greg-Bowers tradition (shown in such language as the "holy grail of authorial inscription" [p. 12]). Unfortunately the essay does not allow for the coexistence (to say nothing of the mutual enrichment) of multiple historical interests; it therefore ends with the prediction that "the day of the critical, or eclectic, edition . . . has passed or is passing" (p. 23), rather than with a recognition of what such editions can contribute to an understanding of the past.[61] The following year Philip Cohen (who had edited Devils and Angels in 1991) brought out another collection of essays,[62] prefaced by an "updated homily on textual instability" (p. 146); and this piece, even more than Hill's, is a repository of the phrases that have become standard in criticisms of intentionalist editing. He says, for example, that eclectic editing "privileges from the start a final authorial and single-text orientation" (p. 134); that there has been "a
In contrast to the many essays that restate in the same clichés the differences between the old and the new textual theories, Morris Eaves has produced a piece that rejoices in textual instability in a genuinely fresh way. His essay for a 1994 anthology entitled Cultural Artifacts and the Production of Meaning: The Page, the Image, and the Body (edited by Margaret J. M. Ezell and Katherine O'Brien O'Keeffe) does not mimic the many recent papers on a socialized approach to texts but instead offers an eloquent argument for the dignity that texts acquire in their public reappearances and reshapings over time.[63] His line of reasoning is linked to Joshua Reynolds's respect for the forms of works that evolve through consensus, for (in Eaves's words) "the process of history, including, at least potentially, the public powers of invention that are only temporarily vested in individual artists" (p. 89). He is thus able to defend the proposition that (in contrast to an editorial view of texts as corrupt) "everything that is, is already right" (p. 85) — or, as he later restates it, "whatever is may be, if not right, then on its way to becoming right" (p. 945). His position rests on "the granting of legitimacy to the audience, and hence to its editorial decisions" (p. 97), which include, in his comprehensive vision, all possible editorial stances. Intentionalist editing by scholars, like every other act of handling a text, fits into the framework, for it is a response by one segment of the audience (which can,
One additional point that Eaves's discussion naturally leads to (though he does not deal with it) is the puzzle of determining what "whatever is" consists of, and the relevance of determining it for anyone interested in history. Reynolds's sanguine belief that what survives is what people wish to survive may possibly be true in the short run, but the long history of textual transmission is full of instances in which lost texts would be eagerly seized and revered in another era. Even in less extreme situations, chance plays its role along with deliberate choice in affecting the forms of texts that are passed along. Nothing can alter what happened, whether or not it was for the best, but readers interested in history (as some scholarly editors are) must recognize the relevance of texts that were no longer extant in physical form at a particular time, as well as those that never existed in physical form. These considerations certainly fit within Eaves's scheme, for they reflect the interests of a portion of an audience, and the products of that interest, which are attempts at reconstructed texts, become available to influence what happens in succeeding generations. Thus when Eaves says that "whatever is may be . . . on its way to becoming right," he recognizes that change may be necessary to produce rightness for a given audience; and his title question, "Why Don't They Leave It Alone?", can only imply a criticism (rather imprecisely) of certain attitudes with which changes are made and does not call into question the activity of making changes, which has always been part of the public accommodation of texts. I mention these points only to clarify the way in which intentionalist editing is encompassed within the ongoing process and to emphasize the fact that the materials of history are not exclusively physical objects. But Eaves cannot be expected to comment on every intricacy of a complex subject, and his argument as it stands is a subtle and sensitive exposition of "the shaping influence of socialization, collaboration, and historical processes" (p. 88). It provides a fitting climax to this survey of anthologies, for it is one of the few pieces in all of them that truly mark an advance in thinking. But the anthology in which it appears will not be the last anthology on textual matters: at least three more are scheduled to appear during the second half of 1995 — one on German and one on French approaches, and one that may well be more important than all the previous ones, D. C. Greetham's Scholarly Editing: A Guide to Research, containing surveys of the textual traditions in some thirty literary fields.[64]
III
In addition to the sizable body of writing that appeared in anthologies during the early 1990s, there are of course other noteworthy articles that were published in periodicals. One of them, Ann R. Meyer's 1994 contribution to Studies in Bibliography ("Shakespeare's Art and the Texts of King Lear," 47: 128 — 146), can serve to focus some of the issues raised by the current concentration on versions and the associated disparagement of eclectic texts. Indeed, it provides a rare model of open-mindedness on this subject. Meyer goes to the heart of recent debates by tackling King Lear, which in the past decade or so has become the classic ground for demonstrating the importance of preserving distinct authorial versions and of refraining from conflating them into eclectic texts. One of the best-known anthologies of the 1980s, The Division of the Kingdoms (edited by Gary Taylor and Michael J. Warren, 1983), set forth the arguments for two versions of Lear, which resulted in the presentation of two separate texts in the 1986 one-volume Oxford Shakespeare.[65] Meyer approaches her independent re-examination of the evidence with the sensible view that "conflation on the one hand and a presentation of different versions on the other are not mutually exclusive alternatives" (p. 130). She recognizes, in other words, that authorial versions cannot be equated with the texts of documents and that, if such versions are what one is interested in, they must be constructed through critical editing. (Early versions are in this respect no different from late or "final" versions.) Armed with this unbiased view of both textual instability and eclectic editing, she finds that — in connection with two key passages — a "judicious consolidation" (p. 145) of the first Quarto and Folio texts is necessary in order to approach an authorially intended text.
Her skillfully presented argument begins with two instances where the Quarto readings in the outer forme of sheet G exist in two states, and she shows in each instance that the "corrected state" (Qb) is simply erroneous and that the uncorrected state (Qa) and the Folio reading, though different, derive from a single version and do not reflect separate versions. Thus "crulentious" (Qa) was altered to "tempestious" (Qb) during the printing of the Quarto, but the Folio "contentious" is the word (as paleographical analysis shows) that was originally misread as "crulentious" (leaving "tempestious" as a compositor's guess at correcting
Having established this pattern, she then considers two major differences between the Quarto and Folio texts: the omission in the Folio of the 31-line mock-trial scene, and the considerable substantive variation in the final lines of the play. As to the first, she argues that the omission, without other accompanying changes, leaves many loose ends and thus is difficult to regard as authorial revision. In the second, the peculiarities in the Quarto text result from the compositor's need to compress the passage to make it fit on the recto of the last leaf, since the outer forme (containing the blank verso of that leaf) had already been printed. Even if there are two versions of Lear, therefore, both of them (according to Meyer and in opposition to the view of the 1986 Oxford editors) would have to contain the trial scene from the Quarto and the concluding scene from the Folio.
The broad importance of Meyer's article, beyond its significance for the Lear debate, is that it sets forth with great clarity several concrete illustrations of the necessity for eclecticism in intentionalist editing, whether or not distinct versions are involved. This point should be accepted as common sense even when stated in the abstract, without accompanying examples; but the persistence with which writers on textual criticism seem to regard eclecticism as antithetical to an understanding of textual instability suggests that articles like Meyer's are needed and need to be read widely. Although she finds no evidence of Shakespeare's revision in the passages she examines, she is open to the possibility that such evidence exists elsewhere in the play. "My argument," she says, "does not reinforce the concept of a 'definitive' or ideal authoritative text, nor does it contradict the concept of the text as a product of many influences, including the possibility of authorial revision" (p. 131). She simply recognizes that, if one is interested in authorially intended versions, one must be alert to "non-authorial influence" (p. 130) in any
The idea that versions cannot be identified with the texts of documents is of course based on a particular view of the ontology of literature (that literary works are intangible); and one should not be surprised that the recent interest in textual versions and instability should have been accompanied by attempts to define the nature of verbal works. The essays by McLaverty and Grigely commented on above are two significant instances. But the most ambitious (indeed, rather self-consciously ambitious) effort of this kind is Peter L. Shillingsburg's essay in the 1991 volume of Studies in Bibliography, "Text as Matter, Concept, and Action" (44: 31 — 82). Shillingsburg has been thinking about the relations among works, versions, and documents since the early 1980s and has published a number of illuminating discussions of them.[67] The long 1991 piece, effectively building on this foundation, offers a "taxonomy" or "anatomy" of texts, resulting in a "partial nomenclature" for textual criticism (p. 46). It tries to address the central problem of how intangible and material texts are linked, and it shows that one can recognize the historical existence of "Conceptual Texts" (texts held in the mind) and still be open to the whole range of possible emphases in the study of texts. (And a "Conceptual Text," Shillingsburg explicitly says, is "not a Platonic ideal" [p. 51].)
For Shillingsburg, a work is a "literary entity" that is "manifested in and implied by the material and linguistic forms" of all the texts that can be thought of as its versions (p. 81); "the Material Text," in other words, "is not equivalent with the Work but is instead merely a coded representation or sign of the Work" (p. 56). Versions, in turn, are not "facts to be discovered" but rather "concepts created . . . by readers [including editors] as a means of ordering (or as justification for valuing) textual variants" (p. 73).[68] In Shillingsburg's terminology, the text of an
Shillingsburg's thoughtful performance is not without its flaws, however. A basic one, in my view, is its fuzziness about how readers' responses relate to the concept of "work." When he says, "it is not the Work itself that is known through the Material Text but the reader's reconstruction of the Work that is known" (p. 58), the idea of "the Work itself" is apparently something associated with the creators or producers of it and inaccessible to its audience. Later, when he says, "the reader becomes the 'functional authority' for the Work and its Versions" (p. 74), the word "functional" perhaps makes this statement consistent with the earlier one; but it occurs in a section entitled "The Reader as Author" (pp. 72 — 74), which also contains this sentence: "The concept of Work and, even more so, the concept of Version depend on Reception Performance just as much as on Creative Performance" (p. 74). And another statement one page later, in the concluding section, goes farther: "the crucial act in relation to a Work of literary art is not writing, or publishing, or editing it, but reading it" (p. 75). There appears to be some indecision reflected in these statements, though the end of the article leaves the impression — quite properly — that "the Work itself" encompasses the reader's, as well as the author's and the publisher's (and other producers'), conceptions of the work.
Even so, the matter is not left entirely clear, for the point about the reader having "functional authority" is followed by the opinion that "ideally the reader should have ready access to the evidence that would fully inform his or her decisions," as in "scholarly editions that foreground rather than submerge the evidence for Versions" (p. 74). But to suggest that readers need historical evidence in the exercise of their authority is to assume that they have historical interests and that their function is to "decode a Work" (p. 75) — that is, to solve a puzzle by finding a pre-existing answer. This approach limits the range of "Reception Performances" and thus restricts readers' authority. I think it would be more profitable to state the authority of the reader somewhat differently. To say that readers' authority is "functional" is in fact to say that it is absolute; and it is indeed absolute, in two senses. First, texts, like everything else, can be known only through the perception of observers. Since it is a given of human existence that what seems to be external to the individual can be apprehended only by subjective consciousness, any discussion of readers' responses should take this point for granted and move on from there. Thus a second kind of authority that readers have is that they are free to decide what kind of interest they wish to have in any text, and their interest may or may not be historical. They can respond to a text and its physical setting as an independent entity, or they can experience it in the context of its relation to other texts and
Another example or two of disappointing features in Shillingsburg's essay may be mentioned. One is his superficial comparison (as it seems to me) of speech acts with what he calls "Write Acts." Usually he finds them so different that "conclusions about speech seem simply inapplicable to writing" (p. 44); but by not referring to oral renditions of literature as speech acts, he neglects some of their similarities (as on pp. 43 — 44, 60 — 61). There are also several false notes in the fifteen pages of introductory comment (devoted in part to characterizing the temper of the times and the recent focus on textual instability and a social approach to texts), such as defining textual criticism in the first paragraph as an activity that concentrates on "removing" error and "purifying" texts,[70] or implying that textual critics have traditionally indulged in "a nostalgic reactionary hope to 're-establish' or 'restore'" texts (p. 35). It is particularly unfortunate that the following sentence appears on the second page, where it can cast a shadow over all that follows: "If textual criticism and scholarly editing are to provide texts and insights that are valuable to literary criticism, they must be conducted in the light of what literary critics find valuable to do." This statement seems to reinforce the old notion that literary criticism and textual criticism are distinct and that the latter is a servant of the former. Shillingsburg himself, both in this essay and in earlier ones, has made clear that he understands editing to be a critical activity; yet he persists here in separating the formulation from the criticism of texts, as when he says that the act of reading has two parts, "the construction of a Reception Text" and "the interpretation of and response to it" (p. 62).[71] In fact the general drift
Another scholar who has been turning his attention to "the boundaries and modes of existence of the literary work" (p. 3) is Paul Eggert, who delivered papers relevant to this subject in August and October 1992. The first of them — published second, in the seventh volume of Text (1994) — is "Document and Text: The 'Life' of the Literary Work and the Capacities of Editing" (pp. 1 — 24), which acknowledges the helpfulness of Shillingsburg's essay. And he is like Shillingsburg in the comprehensiveness of his outlook, in the open-mindedness of his recognition that all moments in the history of a work are worth studying as part of the whole. He emphasizes "the readership's participation in the work" (p. 7) and the corollary of that concept, the idea that the "reception of a work is recognized as part of its constitution" (p. 10) and therefore that the boundary of a work encompasses its "continuing life" as successive generations of readers make it over. But he also recognizes that "the initiating inscriptional acts will always be of crucial interest" and that "to ridicule the link between author and work as mystical is making less and less sense" (p. 14). Indeed, he questions "whether the attempts in recent years to clarify editorial principles by polarizing the authorial
Basic to such a theory, according to Eggert, is "the distinction between text and document" (p. 17), and thus ontological considerations are integral to the practical task of editing.[74] In contrast to Shillingsburg's elaborate scheme for dealing with the relation of the tangible and the intangible, Eggert's approach is refreshingly simple — but is, I fear, overly simple. For him, it can be "a powerful source of clarification" to distinguish the "document" (or "material object") from the "text" (or "textual meaning"); he thinks of them as two "levels," with the text "raised" from a documentary base. "'Text,' under this dispensation," he says, "requires the socialized reader's engagement in the raising of meaning from the document" (p. 2). This statement turns out to signify less than it first appears to, after one reads the footnote attached to it: Eggert says he is leaving aside the meanings conveyed by type design, layout, paper, binding, and so on and is focusing on "the document considered as the physical inscriptions it bears."
Is he therefore only making the commonplace observation that strings of letterforms do not mean anything in themselves and that their linguistic meaning depends on a social convention? The point is of course correct, but how far does it get us? It does not seem to touch the more interesting question of the way in which different readers, all of whom understand how to translate the physical marks into words of a language, attach different meanings to the same documentary text and form different opinions as to which variants (or emendations) are more appropriate. This behavior of readers (including not only editors but authors reading their own work) exists a step beyond the level of recognizing a language. The result of Eggert's "dispensation" is to collapse two stages into the single concept of "text," to no advantage that I can see, and with the disadvantage of being less precise. Eggert complains that McGann's "linguistic text" is "not just there on the page for all to recognize" (p. 2); but in an important sense it is — the page displays a physical "text" made up of inked marks that conform to the notation scheme for a particular language. If we wish to talk about the "recognition" of those marks, we are talking about something very different from the act of reading that builds on the knowledge of what those marks signify.[75] When Eggert — in line with his admirable aim of seeing production and reception as a whole — says that the document is "the tangible link between writer and reader" (p. 11), he is actually presupposing that the marks put there by the writer (or scribe or compositor) follow a system that can be identified by a reader. Eggert is perhaps right to believe (along with many others) that the most fundamental problem in textual criticism is to understand clearly the connection between texts in the mind (whether the writer's or the reader's) and texts on paper; but the way he differentiates them in his concepts of "text" and "document" does not seem likely to promote such clarity.[76]
Eggert's other paper (which uses "text" and "document" in the same way) attempts to clarify the nature of literary works by comparing and contrasting them with paintings ("Editing Paintings/Conserving Literature: The Nature of the 'Work,'" Studies in Bibliography, 47 [1994], 65 — 78). It pursues, intelligently but unremarkably, the connections between verbal and visual works, showing that in both cases the work is "a creature of our own conventions" (p. 67), with "no unchanging existential anchor," and that "conservation and scholarly editing must be understood as interpretive rather than scientific or technical activities" (p. 76). Eggert also recognizes the "significant difference" in the two situations: that literary editors, unlike conservators of paintings, need not alter surviving artifacts in the course of carrying out their work. None of this is surprising or questionable; nor is the conclusion (for which the analogy with painting is not actually needed) that scholars editing a literary work are not engaged in "releasing" it but are "participating in its ongoing life" (p. 77). The last paragraph of the piece, however, will raise a question in many readers' minds — an epistemological question that is inherent in much recent theorizing about texts but one that is brought up more explicitly than usual by Eggert's wording. His "phenomenological view of the work" (which "would abandon any belief in the work as an ideal thing") causes one to realize, he says, that theoretical debates are "about our conventions for understanding the work, not something inherent in it — for there is no it, as work of art, independent of our understandings" (p. 77). The last clause (following the dash) is incontestable, and yet it undercuts the rest of the sentence, for it describes a universal condition, applicable to everything. Any reader might be excused for assuming that this point could be taken for
A more probing treatment of the relation of visual arts to literature (though it ends with a similar quandary) is Joseph Grigely's "Textual Criticism and the Arts: The Problem of Textual Space" (Text, 7 [1995], 25 — 60). As usual, Grigely is stimulating and well-informed and is refreshing in his choice of illustrations.[77] His account of the visual arts provocatively asserts that one should not equate "the artwork and the art object" (p. 43) and that so-called reproductions of paintings are like editions of verbal texts in extending the "space" occupied by the works: "To speak of the original as being privileged in art has some of the same overtones as championing the author's intention in literature" (pp. 45 — 46). This opinion is inherent in his general position that all texts are "caught in a field of cultural forces" (p. 58) and that "there is no detachable space in which texts manifest themselves" (p. 59); works have no boundaries, for "our perception of the 'work' is an extension of the historical and physical space inhabited by a specific text of that work" (p. 44). Trying to fix boundaries is like "drawing circles in water" (p. 34).
This approach equates works with all other events that can have mental repercussions: "the problems of textual space in literature . . . appear in most human contexts — not just the arts, but as a part of human relations in general" (p. 42). One begins to wonder whether such all-inclusiveness is not self-defeating for the essay. Practically everyone understands that the experience of a text, like all other experiences, is colored by one's entire previous life and that it will continue to reverberate in one's mind, affecting all experiences afterward. An essay that says something like this does not mark any advance in thinking, though perhaps to reiterate the point as engagingly as this essay does is its own justification. Yet one cannot help feeling that analytic discourse involves making distinctions; we know that everything is part of human experience, but it may be enlightening to attempt to set boundaries between one kind of experience and another. It is not merely "nominalist" (to use a favorite term of Grigely's) to believe that discriminations can be helpful in developing
One might think that philosophers could help literary scholars with these matters, but in general their treatments of such topics as artifacts and texts do not bear directly on the concerns of textual critics, frequently because of their unfamiliarity with the point to which textual critics have already carried the discussion. Mark Heller's The Ontology of Physical Objects: Four-Dimensional Hunks of Matter (1990) takes up a subject that should be of intense interest to all who are concerned with any of the arts (tangible or intangible), for they all make use of physical objects. It is not Heller's purpose to apply his findings to the arts, and the details of his argument will be of limited use to textual scholars; but his general thesis should be borne in mind by all who deal with artifacts. His main point (concisely expressed in his subtitle) is that objects have a fourth dimension, time, and that a physical object should be defined as "the material content of a region of spacetime" (p. 4).[79] The role of time in defining objects has not been absent from textual criticism (the best recent example being McLaverty's essay discussed above), but Heller offers a careful restatement of the point.
Less helpful, despite the potential of its topic to be more so, is Gregory Currie's "Work and Text" (Mind, 100 [1991], 325 — 340). It may be reassuring to learn that a philosopher's technical arguments support the position that works and texts are very different things, but many of Currie's points will sound naïve to textual scholars. His equation of a "definitive text" with the text intended by the author (p. 326) is not in itself the problem (since the particular agent selected is not crucial to the argument); what is more troublesome is his simplistic separation of language and interpretation, as in his concluding discussion, where interpreting a text is stating "the meaning of its constituent words and sentences as they are given by the conventions of the language," and interpreting a work involves "the explication of plot, the limning of
Another group of writings that contribute little to conceptual thinking about textual criticism consists of those dealing with the "politics" or "rhetoric" of editing. One might at first assume that there is no reason why they should; yet the frequent use of these terms to refer to the subjectivity or the attitudes embedded in editions (as in two anthologies with "Politics" in their titles, the Spadaccini-Talens The Politics of Editing [1992] and Roberta Frank's The Politics of Editing Medieval Texts [1993]) is directly connected with the theoretical emphasis on the social construction of texts, and in this sense the "politics" of editing cannot be seriously discussed without pursuing theoretical matters. But even the most substantial of these pieces, D. C. Greetham's "Politics and Ideology in Current Anglo-American Textual Scholarship" (Editio, 4 [1990], 1 — 20) does not (and does not attempt to) make a substantive contribution to textual theory; instead it is historical and descriptive, first outlining the professional organization of the field ("the organisational manifestation of ideology" [p. 11]) and then the "challenges" to the "Greg-Bowers ideology" (p. 19). At least it has the value of a convenient overview,[83] but most of the writings on "politics" are much slighter: witness the contributions to a panel discussion, at the 1991 Society for Textual Scholarship conference, on "The Politics of Editing," which dealt with matters of funding, attracting talented scholars to the field, and finding receptive publishers.[84] Anyone who wishes to follow the
IV
What can be regarded as a contribution, in any field, is likely to be a different angle of vision, an unaccustomed way of perceiving something. Rarely is an entirely new idea formulated, but old ideas can usefully be given new emphases or be placed in new contexts. The recent dominance of textual instability as a subject for writers on textual criticism is valuable because it directs attention to an aspect of textuality that has not been adequately explored in the past. It has of course been recognized by everyone who has ever thought about texts: the changeability of texts over time is the basic fact that gave rise to such a field as textual criticism and has been the grounding for all editorial activity. But the fact of instability was taken as the starting point, as the essential condition within which one had to work, not as a particular focus of interest in its own right. There has always been an implicit understanding that individual versions are of interest, and this understanding has often taken explicit form in the construction of apparatuses and the publication of facsimiles; but the primary concern of textual scholars before the last few decades was authorial intention, and the primary editorial activity was presenting texts (however arrived at) that reflected authorial intention. Thus the recent interest in the process of textual metamorphosis, along with the emphasis on the value of every one of the myriad forms that texts of works take (however mixed — in many of those forms — are the intentions of authors with the intentions of others), is a most welcome development, throwing a spotlight on an area where it had not previously been directed.
Any such development, however, is apt to be accompanied by two kinds of problems. One is that enthusiastic advocacy for the new may involve an unfounded denigration, or an inaccurate characterization, of what went before. The recent writings on textual instability, the social construction of texts, and related ideas have amply demonstrated this point. Given the fact that the new position is characterized by an openness to all the forms of texts produced by the historical process, it is
The other kind of problem often associated with new theoretical insights is that they may leave people at a loss to know how to proceed. Although it is no criticism of a theory to say that its practical implications are unclear, people are nevertheless bothered by theories that seem to lead to inactivity or to an impasse. Many of the essays on textual theory in recent years have lurking within them the implication that editing is an impossible, or unnecessary, activity. If all the variant texts that have existed are worthy of attention as the products of social forces and as the inspirations for readers' responses (and this point is clearly valid), then it may seem that there is nothing for editors to do. This feeling of helplessness is analogous to the aporia of deconstructionist readers, faced with words and texts of indeterminate meaning; and this similarity is not surprising, since recent textual theory springs from the same intellectual milieu that produced recent literary theory. Much of that theory is convincing in its own terms, and it frequently does help us to understand a little better what complex events are subsumed in the concepts of "text" and "reading." But whether or not acts of communication can theoretically be consummated, most of the time we behave as if we are not trapped in private prisons of language. Writing to express thoughts, as well as reading to receive them, will continue to be practiced, simply because they are activities that are congenial to the human mind. Reveling in or despairing over the impossibility of succeeding in these activities is also present in some minds, but it can coexist with the practicing of them. Similarly, scholarly editing will continue to exist because it, too, is a natural activity of mind. It is one of the forms of response to texts: of those readers who make public responses to what they have read, some write essays, some give lectures, and others produce editions. The existence, on some occasions, of a gap between what we do and how we theorize about it does not in any way suggest
Editing is the practical side of textual scholarship, and the recent emphasis in textual theory on the importance of versions calls increased attention to a long-recognized practical problem: how best to present texts and textual evidence, particularly in codex form. The concern with versions has international force behind it; in addition to the many discussions in English (from scholars in most of the English-speaking countries), German textual scholars have long concentrated on versions, and the French now have a flourishing school of critique génétique, emphasizing the pre-publication evolution of texts.[85] Much of this work has been practical, in that it has applied the theory of the significance of versions (set forth in a limited number of general treatments) to the textual history of specific writings; but the results have taken the form of essays as often as (probably more often than) of editions. The presentation in codex form of complete texts of versions has always posed difficult problems, both because of the space they take up (and the consequent unwieldiness of the resulting volumes) and because of the inconvenient process they entail for the detection and comparison of variants. For these reasons, nearly every editor in the past has presented only a single text (whether critical or diplomatic) in complete form and has recorded the variant readings from (and other information concerning) all the extant documents in abbreviated form in an apparatus — either incorporated into the single linear text (and identified with symbols) or appended to that text (at the foot of each page or at the end of the volume). Because the codex form forced this kind of compromise on editors, they have discussed endlessly the questions of selection, arrangement,
Jo Ann Boydston's "In Praise of Apparatus" is particularly eloquent in its account of how the reading of an apparatus is "a stimulating and highly productive intellectual adventure" (p. 10). Her attitude echoes that of a great many other editors who have produced editions containing single eclectic texts with accompanying apparatus: their presentation, in each edition, of only one text in full does not mean that they fail to see the importance of other versions. Some of the recent critics of eclectic editing have assumed that conventional editions of the past reflect a theoretical belief in the primacy of a single "ideal" text for every work (or for most works). But a more realistic explanation for the kind of presentation given to texts and variants in past editions is the constraints of the codex form. Given the effort that most editors have expended on apparatus and the stress they have habitually laid on the importance of lists of variants for understanding the textual history of a work, it is unreasonable to think that they have not understood the significance of versions as part of the process of experiencing a literary
It is true, of course, that the single texts chosen for full presentation by editors (having been forced to make such a choice) do reflect personal judgments as to what will be most useful to most readers. The judgments inevitably involve some mixture of practical and theoretical considerations, but they cannot be assumed to arise from any lack of openness, on theoretical grounds, to multiple texts. At the same time, they may very well signal a belief, on theoretical grounds, in the value of critically constructed (that is, eclectic) texts. Those recent writers who have attributed to intentionalist editors the notion that there is a single ideal text for each work have blurred two separate issues — singleness and ideality. The use of the word "ideal" is itself part of the problem. Sometimes it seems to mean something like "best" or "perfect"; at other times it means "not real" in a physical sense — that is, not extant in a document. The first sense (not often used by critical editors) implies singleness; the second is fully compatible with textual instability. A conviction of the importance of attempting to reconstruct nonextant texts does not entail any concept of the sufficiency or finality of a single text; indeed, critical editors are well aware not only of the lack of finality attaching to any product of critical judgment but also of the fact that any number of past moments (not just the one judged to represent final authorial intention) can be selected for critical reconstruction. Critical editors of the past clearly believed in the importance of critical texts. But their presentation of single texts did not necessarily mean (and in fact was not likely to have meant) that they believed only one text was valid or desirable; it only meant that the option of presenting more texts was not open to them.
Technological developments have now made that option feasible, and editors are rightly excited by the possibilities that electronic presentation offers. The capabilities of word-processing and hypertext programs for textual study have already been the subject of a considerable literature, ranging from (to name two works from the early 1990s) the general theoretical treatment of George P. Landow's Hypertext: The Convergence of Contemporary Critical Theory and Technology (1992) to the detailed practical overview of Charles B. Faulhaber's "Textual Criticism in the 21st Century."[87] In a hypertext edition, one can have
This is not the only requirement that we should have for electronic editions, however. The elimination of space constraints takes care of the problem of the "single" text that was associated with the codex form, but it has nothing to do with the kinds of multiple texts that get presented. Writers on hypertext editions frequently think of the goal as a kind of "archive" limited to documentary texts. The dissemination in this way of manuscript and printed texts surviving from the past is extremely valuable, obviously, just as the publication of documentary editions has always been valuable. But hypertext editions offer great advantages for the presentation of critical texts as well, and one can argue that the potential of the electronic form is not being very fully exploited unless editors' critical reconstructions are included along with documentary texts. Critical texts may be out of favor among many theorists at present, but there will always be scholars and other readers who understand the need for reconstructions of additional historical moments besides those represented in surviving documents. And for such persons, electronic editions offer the possibility of multiple (and linked) critical texts, attempting to show different intentions (those of a publisher, say, as well as those of an author) as they existed at different times.
One of the ways that traditional printed critical editions can be faulted is that, whereas the apparatus was used to record the variants in documentary texts, there was no attempt to use it to show what editorial emendations should be made to produce other critical texts besides the one presented as a full reading text. (This criticism has of course not
Even an electronic edition of the kind envisaged here, with images of primary records, newly keyboarded texts of them, and a range of critical texts, would not satisfy some recent theorists, for two reasons: the reader is still removed from the physical objects that originally conveyed the documentary texts, and the reader is dependent upon the subjective reconstructions of a single editor (or series of editors, no matter how numerous). Neither of these points can be denied, but whether they are grounds for complaint is an interesting question because it involves the most fundamental characteristics of editing as a practical undertaking. The first fact that one must confront in thinking about it is that not all readers are interested in history. There is no reason why they should be, if they do not wish to be or if (whether or not they have considered the matter) they are not temperamentally inclined to give any thought to the past. Such readers can respond to the linguistic and design features of the document (whether paper or electronic) that they have in front of them, and how those features compare with the ones presented to past readers of what might be called the same work is quite properly of no concern to them. Scholarly editing, when it is conceived of as an activity oriented toward historical recovery, is irrelevant to their purposes.
So, in turn, are most of the theoretical debates, for most of them presuppose
Either we are talking about historical approaches to literature, or we are not. If not, then the arguments of theorists that scholarly editing (or any kind of editing, for that matter) creates a barrier to historical discovery and understanding is irrelevant. But if we do wish to be concerned with literature in its historical setting, there is still good reason to believe that recent theoretical arguments critical of scholarly editing are misstated. They do not always say explicitly that there is no future for editing, but the idea is implicit in them: for if all the physical details of documentary artifacts are essential to the historical experience, then facsimile editions are not adequate; and if all documentary wording must be directly encountered by readers, without the intervention of editors' subjective emendations, then editions with critical texts are inappropriate. Each of these points can, however, be stated in a more understanding way; and doing so leads to the conclusion that both kinds of editions are not only inevitable but are desirable and necessary.
Of the importance of artifactual details in the historical understanding of texts inscribed or printed on physical objects, there can be no doubt. Whether or not authors or readers in any given instance regard physical characteristics as part of the text, those characteristics do reveal information about the production of the objects, the social milieu of the text, and the bases for readers' responses. Every visual and tactile detail is relevant, and no attempted reproduction can possibly carry the same historical suggestiveness as the object that survives from the time in the past that is the subject of one's interest.[91] But editors have always
Another category of dissatisfaction that recent theory has found with editorial activity relates to the production of critical texts. The principal objections are that a critical edition conceals the fact of textual instability by presenting a single "ideal" text and that such a text offers an unhistorical conflation reflecting an editor's subjective judgments about authorial intention. In order to focus on the real issue here, we can
Many writers have commented, as Shelley did, on the difference between the idea for a work and the executed forms of the work. Virginia Woolf put it this way:
Accepting the necessity of critical extrapolation from artifacts still leaves open the question of whether the results should be published. That in effect is the question being answered in the negative by those recent theorists who say that readers should decide for themselves how (or whether) they wish to alter documentary texts. Readers will indeed finally make this determination, but they are not all equally qualified to engage in historical reconstruction, which involves knowledge as well as imagination, and they may wish to have the results of specialists' critical thinking. No one would be likely to claim that historians and literary scholars should not publish essays on the grounds that readers should not be told what to think. Similarly, one can scarcely claim that critical editions (which, like historical essays, are the products of systematic efforts to interpret the past) are objectionable because they inflict
Notes
The deficiencies of what is now the stock response to intentionalist editing have occasionally been pointed out. Tim William Machan, for instance, in his review of Paul Eggert's anthology Editing in Australia (1990; discussed below), gives examples of "the factual and logical imprecision" that often accompanies the exorcism of "the deadly, if amorphous, shade" called "Greg-Bowers"; he concludes that "critics ought to try to understand the Greg-Bowers line of thinking before they condemn it, and in the current climate it has become all too easy not to do so" (Text, 6 [1994], 383 — 386).
This is the fifth in a series of articles surveying the period since the middle of the century. The first three, which were originally published in Studies in Bibliography in 1975, 1981, and 1986, were collected in book form as Textual Criticism since Greg: A Chronicle, 1950 — 1985 (1987). (Any reference to these articles in the present piece provides the SB citation first, followed in brackets by the page reference to the 1987 book.) A fourth essay, "Textual Criticism and Literary Sociology," covering the last half of the 1980s, appeared in SB, 44 (1991), 83 — 143. The scope of these pieces is the same as that of the present one, focusing on writings of general theoretical significance in English. I have compiled a comprehensive list of such writings (including many from 1990 — 95 not referred to in the present article), distributed by the Modern Language Association's Committee on Scholarly Editions; the latest revision is A Sixth Interim Supplement (1995) to "The Center for Scholarly Editions: An Introductory Statement" (1977) (1995). (Some of my own writings have been a part of the current debates, but it is not my purpose here to comment on, or reply to, discussions of my work; how I would respond will be clear in any case from what I have said below about the writings of others. I have briefly described my own view of what I have tried to accomplish in footnote 3 of "Books, Canons, and the Nature of Dispute," Common Knowledge, 1.1 [Spring 1992], 78 — 91. A collection of my essays was published in 1990 as Textual Criticism and Scholarly Editing.)
This breadth is both a strength and a weakness: the inclusion of all these topics in a volume with the words "textual scholarship" in its title makes an important point in itself, by suggesting how interrelated with textual study are all aspects of the history of books; but the attempt to cover so much inevitably results at times in oversimplification and unevenness. For a favorable assessment of the book, see James Thorpe's review in Text, 7 (1994), 543 — 546; less favorable evaluations are offered by John Winter, Elizabeth Morrison, and B. J. McMullin in "Symposium on D. C. Greetham's Textual Scholarship: An Introduction," Bibliographical Society of Australia and New Zealand Bulletin, 19 (1995), 47 — 60.
It has been reviewed in Analytical & Enumerative Bibliography, n.s. 4 (1990), 129 — 133 (by Rodger L. Tarr); Review, 12 (1990), 69 — 79 (D. C. Greetham); Review of English Studies, n.s. 42 (1991), 431 — 432 (Peter Davison); and Text, 6 (1994), 359 — 365 (James L. W. West III).
This volume, edited by Joseph Gibaldi, is an entirely separate volume from the one with the same title edited by him in 1981. The contents of the two are completely distinct and by different authors; for the earlier volume, I wrote the essay on "Textual Scholarship."
This volume, scheduled for publication in the fall of 1995, is edited by D. C. Greetham and contains a series of essays describing the history of the editing of many national literatures, including ancient and non-Western writings. My essay incorporates a diagram that aims to show the relationships among the different kinds of editing that are possible. (A still shorter introductory essay of mine is the entry on "Textual Criticism" in The New Princeton Encyclopedia of Poetry and Poetics, ed. Alex Preminger and T. V. F. Brogan [1993], pp. 1273 — 76.)
His death has occasioned some assessments both of his career and of the field during his time: see, for example, my The Life and Work of Fredson Bowers (1993), also printed in SB, 46 (1993), 1 — 154, and the "Fredson Bowers Commemorative Issue" (Second Quarter 1991) of the Bibliographical Society of Australia and New Zealand Bulletin (15: 45 — 104).
One of his last articles — "Authorial Intention and Editorial Problems," Text, 5 (1991), 49 — 61 — begins with the statement, "The purpose of this paper is to avoid generalizations on editorial theory." For his last two (posthumously published) articles, see notes 34 and 86 below.
A similarly exaggerated statement near the end of the book asserts that the "Greg-Bowers scheme . . . ought to have found room for at least some of the elements of collaborative creativity" (p. 199). Stillinger is not the only person who has recently given evidence of a failure to understand the complexity of the concept of authorial intention as ordinarily used by editors in the Greg-Bowers tradition. For example, Margreta de Grazia and Peter Stallybrass end their 1993 article on "The Materiality of the Shakespearean Text" (Shakespeare Quarterly, 44: 255 — 283) by saying that "solitary genius . . . is, after all, an impoverished, ghostly thing compared to the complex social practices that shaped, and still shape, the absorbent surface of the Shakespearean text." A concern with authorial intention can seem "impoverished" only if one fails to recognize in it the psychological subtleties (and, indeed, social relationships) inherent in all human events. (This article shows further striking confusion in its linking of "solitary and unitary authorship" with a rejection of multiple versions [see p. 276]. The authors' failure to sort out clearly the issues involved is perhaps foreshadowed on their first page in the tendentiousness of their remarkable choice of the word "resentment" in their description of recent editorial history: "One of the most evident results of the multiple-text issue has been mounting resentment toward the editorial tradition.") Another article that displays an extreme misunderstanding of the motivations and goals of intentionalist editing as usually practiced is Howard Marchitello's "(Dis)embodied Letters and The Merchant of Venice: Writing, Editing, History," ELH, 62 (1995), 237 — 265. Marchitello inaccurately associates intentionality with the "fiction of the wholly autonomous author" and the "production of texts outside or beyond both culture and history" (p. 237); the simplistic view of history implied here is also manifested in his repeated assertions to the effect that "unediting" (or "a theory of radical unediting" [p. 242]) is necessary to "return the text more fully to history" (p. 241; cf. pp. 259, 260).
Indeed, an ardent champion of the intentions of single authors, Donald H. Reiman, goes so far as to believe that what authors expected others to do is a part of their intention. In "Public and Private in the Study of Manuscripts," Text, 6 (1994), 49 — 62 (which is a summary of the book based on his 1989 Lyell Lectures, The Study of Modern Manuscripts: Public, Confidential, and Private [1993]), he unequivocally speaks of "the absurd view that the unitary author does not exist" (p. 52). At the same time, he thinks that the vexed distinction between authorial intention and social process will disappear if one recognizes that modern manuscripts were intended only as "way-stations" to published texts, thus sanctioning certain kinds of alterations made in the publication process. But his idea that the concept of "public" manuscripts (that is, manuscripts containing texts intended by their authors for public dissemination) entails the merging of expectation with intention oversimplifies a highly complex matter. It is surprising that Reiman, as a person interested in authors as individuals, would not wish to distinguish what authors personally preferred from what they expected and accepted; my point is not that one is necessarily of more interest than the other but that they are two separate interests, both important as history. When I refer, in the text above, to intentionalist editors recognizing collaborative intention, therefore, I am not thinking primarily of editors who take a position like Reiman's; for it is not necessary to blur the concept of intention, by including expectation, in order to recognize that authorial intention in the strict sense (meaning what an individual author wished to place in the text at each point) can be expected to include outside influences. (For a thoughtful discussion of Reiman's book, see D. C. Greetham, "Getting Personal/Going Public," Review, 17 [1995], 225 — 252.)
Parts of which had previously appeared in his article "Multiple Authorship and the Question of Authority," Text, 5 (1991), 282 — 293.
My earlier essays in this series (see note 2 above) have repeatedly dealt with these questions, and the arguments need not be repeated here. As examples of Stillinger's simplistic summaries, one might note the following: "Adherents of the Greg-Bowers dogma sometimes appear overly rigid . . . settling the question by general rule rather than by an assessment of particular circumstances" (p. 197); "The wording of the first edition or manuscript could be emended . . . but the punctuation, spelling, capitalization, word-division, and paragraphing would remain those of the first edition or manuscript" (p. 196); "it is not clear to everybody's satisfaction why final versions or latest substantives, merely because they are latest, should be considered more authoritative than any other that carry the writer's authority" (p. 197). Stillinger is not alone, of course, in uttering such distortions. Similar points are made, for instance, by Allan C. Dooley in the final chapter ("Textual Change and Textual Criticism") of his Author and Printer in Victorian England (1992) — a chapter reprinted, in somewhat revised form, as "Varieties of Textual Change in the Victorian Era," Text, 6 (1994), 225 — 247. He asserts that the Greg-Bowers "school" is concerned with "an author's initial intentions": for this "camp," "earlier is always better, whether we are considering revisions of unknown origin or a genuine authorial second version" (p. 171). Not only does this position depend too much, he believes, on "chronology" (p. 173); it also rests on "the romantic concept of inspiration," in which the "artistic impulse can never be adequately realized in words" (p. 171), and it "tends to elevate the textual critic's judgment over the author's in matters of revision" (p. 173). Bizarre as this picture of Greg-Bowers editing is, Dooley goes even farther to compound the confusion. He describes a "rival position" ("which has no agreed-upon name") that favors "an author's final intentions about a text"; and then he sets forth his own view, urging editors to seek "that text which most fully embodies the author's best, most complete, most successful effort to get the work right" (p. 174). What this can possibly mean as a procedural directive (especially in the absence of any reference to emendation) is not clear, and the reader's puzzlement can only increase upon reading that "This approach properly respects an author's proprietary rights over a text, while not necessarily taking all authorial revisions as improvements" (p. 174). Dooley's subject did not require him to comment on editorial matters, and he says in a footnote, "I will forgo any attempt to integrate this chapter's generalizations into current textual theory" (p. 170); his book would have been stronger if he had followed this resolve. His book and Stillinger's are alike in at least one respect: neither calls for discussion of textual theory, and — given the way such discussions turned out — both would have been better off without them.
A similar statement appears in Stillinger's later book, Coleridge and Textual Instability: The Multiple Versions of the Major Poems (1994): "author's intentions are in general unrecoverable apart from the texts that authors produce" (p. 135). One should remember, however, that authors may well have produced texts not now extant.
The place of criticism in editing is not made clear here in any case. Stillinger continually refers to "interpretation and editing" as two activities (e.g., p. 183) and organizes his final chapter on the basis of this division. A fuller recognition that editing (like all historical scholarship) is itself criticism might have resulted in a less naïve concept of historical "availability."
He thinks he is making a valid criticism when he says, "The theorists do not treat facts as if they were speculation, but sometimes they treat speculations as if they were fact" (p. 201). The distinction between fact and speculation, however, is not self-evident; historical inquiry has no choice but to treat speculation as fact, because facts are speculations that informed observers agree to accept until they are persuaded by a contrary argument. (See my "Printing History and Other History," SB, 48 [1995], 269 — 289 [esp. pp. 283 — 286]. I have also commented on the historical nature of critical editing in Libraries, Museums, and Reading [1991], esp. pp. 29 — 31.) Earlier on the same page, Stillinger naïvely contrasts scientific theory with "interpretive and editorial theory" by claiming that the former "is, sooner or later, verifiable."
Even the chapters setting forth the case studies are at times marred by such intrusions as the following: at the end of the chapter called "Pound's Waste Land," we are told that "The Waste Land, if it were perceived to be a jointly authored poem, would inevitably become a lesser work than it is now taken to be" (p. 138). Similarly, the account of the textual history of Sister Carrie would be stronger if it were not tied to a shortsighted criticism of the Pennsylvania edition as "an essentially fanciful construct" (p. 161).
I have described and analyzed it in — among other places — SB, 39 (1986), 19 — 27 [127 — 135], and 44 (1991), 99 — 112. (Another of the best-known advocates of a social approach to bibliographical and textual study, D. F. McKenzie, also made a significant statement during the early 1990s: his centenary lecture for the Bibliographical Society, "What's Past Is Prologue": The Bibliographical Society and History of the Book [1993]. I have commented on it in "Printing History and Other History" [see note 15 above], pp. 282 — 283 [note 27], 287 — 288.)
It is strange, given McGann's emphasis on this point, that he begins his introduction this way: "Both the practice and the study of human culture comprise a network of symbolic exchanges. Because human beings are not angels, these exchanges always involve material negotiations" (p. 3).
McGann claims that the world reflected in his approach "comes into focus when we ask James McLaverty's provocative question: 'If the Mona Lisa is in the Louvre in Paris, where is Hamlet?' In this world, time, space, and physicality are not the emblem of a fall from grace, but the bounding conditions which turn gracefulness abounding" (p. 9). The question (which is not really McLaverty's, of course, but a cliché of ontological discussions in aesthetics) can be useful if it causes people to recognize that some arts use tangible materials and some do not; the realization that authors' intentions may not be adequately represented in existing physical documents follows as a logical corollary and has nothing to do with a view of the material world as fallen from grace.
Such as Fredson Bowers's statement, in his edition of William James, that the apparatus is of "equal ultimate importance" with the main text because it shows "the progress of James's thought from its earliest known beginnings to final publication in journal and book, and continuing to annotation in his private copies" (Pragmatism [1975], pp. 182 — 183). For further discussion of this point, as well as the way in which the printed forms of texts in pre-electronic days limited the flexibility editors had in presenting multiple texts, see the last pages of the present essay; see also my "Critical Editions, Hypertexts, and Genetic Criticism," forthcoming in a special issue of Romanic Review (86.3) deriving from an April 1994 conference, "From Manuscript to Text: Genetic Criticism and Literary Studies," sponsored by the Institut des Textes et Manuscrits Modernes of the Centre National de la Recherche Scientifique, Paris, and the Department of French and Romance Philology, Columbia University.
At the end of the book McGann says that his "theory of the radical instability of the material and conceptual 'text'" allows one "to imagine the possibility of reliable knowledge," because knowledge emerges through the study of successive textual engagements, each of which "localizes human temporalities" (pp. 185 — 186). It is surprising that he does not recognize how intended texts fit into this historical process.
These essays are grouped in two sections. The first, entitled "The Garden of Forking Paths," consists of four essays: "Theory, Literary Pragmatics, and the Editorial Horizon," from Devils and Angels: Textual Editing and Literary Theory, ed. Philip Cohen (1991), pp. 1 — 21 (where the title is simply "Literary Pragmatics and the Editorial Horizon"); "What Is Critical Editing?", from Text, 5 (1991), 15 — 29; "The Socialization of Texts," from Documentary Editing, 12 (1990), 56 — 61; and "The Textual Condition," from Text, 4 (1988), 29 — 37 (which I have briefly commented on in my 1991 SB piece [see note 2 above], footnote 35). The second section of the book is an extensive case study of Ezra Pound ("Ezra Pound in the Sixth Chamber"), made up of three chapters, the first previously published — "How to Read a Book," Library Chronicle of the University of Texas at Austin, 20 (1990), 13 — 37 (this journal number was also published separately: see the text at note 40 below). Two of these chapters were the subject of commentary in the journal issues in which they originally appeared. Following the 1990 piece in Documentary Editing came Hershel Parker's "A Position Paper on Authorial Intention and the Socialization of Texts" (pp. 62 — 65), in which he laments the absence of what he calls "old-school history" in the social approach, which he feels has neglected "the aesthetic and commercial principles of the participants in the socialization process." And following the 1991 piece in Text came T. H. Howard-Hill's "Theory and Praxis in the Social Approach to Editing" (pp. 31 — 46), which includes a response that many people have felt when confronted with the idea that every edition is validated by history: "if this is true, . . . then it is no longer possible to edit works at all" (p. 41). Among Howard-Hill's concerns is doubt as to how the "bibliographical" (or physical) features of a documentary text (which McGann would not wish to separate from the "linguistic" text) can be "encoded" into an edition, unless facsimile reproduction is used. In "A Response to T. H. Howard-Hill" (pp. 47 — 48), McGann says that "technology is making it possible for us to see (theorize) more of the literary work's signifying dimensions" and to "translate such theoretical knowledge into practical editorial work." Although the question could have been more acutely framed, the response is decidely unhelpful in seeming to claim that technology makes theory possible. (For another criticism of McGann, see note 73 below.) McGann also commented on Rodger L. Tarr's review of The Textual Condition (Analytical & Enumerative Bibliography, n.s., 7 [1993], 3 — 12) on the two pages following that review. He properly points out that "the emergence of electronic text has provided scholars with a tool for overcoming the limits of the codex" (p. 14); but his opinion that "The theory of copytext editing is a function of the codex" (p. 13) fails to differentiate between the reporting of textual information (which is indeed restricted by the codex form) and the editorial theory brought to bear on the textual situation (which is not so restricted). "Copyright editing" does not prohibit the presentation in full of multiple texts (a point I comment on more fully in the last section of the present article).
He does not add to the clarity of his discussion by using "linguistic" to label one of the two strands of an "event" that itself is called "linguistic."
This editorial tradition, it should be remembered, is the one that developed analytical bibliography and thus gave great attention to physical clues that reflect production history; but such uses of physical details are of course very different from the uses that McGann is talking about.
McGann has devoted a subsequent book to this subject: Black Riders: The Visible Language of Modernism (1993) develops the thesis that "twentieth-century poetry in English is a direct function and expression of the Renaissance of Printing that began in the late nineteenth century" (p. xi). A number of recent studies have explored, in various ways, the role of book design in the production and reception of literature; some examples are David Foxon's Pope and the Early Eighteenth-Century Book Trade (ed. James L. McLaverty, 1991), Tom Conley's The Graphic Unconscious in Early Modern French Writing (1992), Edward A. Levenston's The Stuff of Literature: Physical Aspects of Texts and Their Relation to Literary Meaning (1992), David McKitterick's "Old Faces and New Acquaintances" in Papers of the Bibliographical Society of America, 87 (1993), 163 — 186, Evelyn B. Tribble's Margins and Marginality: The Printed Page in Early Modern England (1993), and the anthology edited by Margaret J. M. Ezell and Katherine O'Brien O'Keeffe (and commented on below), Cultural Artifacts and the Production of Meaning (1994). Joseph Viscomi, in his remarkable Blake and the Idea of the Book (1993), includes a substantial discussion of "Editing Illuminated Books" (pp. 151 — 183), which should not be overlooked by editors of other authors, since it presents a perceptive and reasonable approach to the problems posed by texts that combine the verbal and the visual. (See my review in Nineteenth-Century Literature, 49 [1995], 534 — 537.)
McGann uses these adjectives to describe both Fredson Bowers's work and "the formal and thematic hermeneutics that cut a parallel course in interpretive studies" (p. 22). It is a gross misconception of both the Greg-Bowers tradition and the New Criticism to regard them as "ahistorical."
These words occur in the description of the first of "three basic choices" for editing Rossetti's The House of Life. The second is to offer diplomatic texts of "a series of versions," and the third is "to construct a text which would represent that textual evolution" (pp. 29 — 30). It is important to recognize, however, that these three choices are really only two: constructing a critical text or recording documentary texts. Whether the latter is handled through the printing of separate texts or the production of a genetic text is a matter of reporting; they are both subcategories of the same basic approach.
Studies in Bibliography also takes the form of annual volumes, but they are not exclusively devoted to textual matters. (The German annual Editio is so devoted but rarely includes articles in English.)
These figures are based (with necessary adjustments for differences in scope) on the list of anthologies that I included in "A Concise Selection from the Literature of Textual Criticism," which appears as Supplement 4 to my Introduction to Bibliography: Seminar Syllabus (3rd ed., 1994) and as Appendix I to A Sixth Interim Supplement (1995) to "The Center for Scholarly Editions: An Introductory Statement" (1977) (1995). My list may of course not be complete. (For references to some of the French and German anthologies, see note 85 below.)
They were Bibliography and Textual Criticism, ed. O M Brack, Jr., and Warner Barnes (1969); Art and Error: Modern Textual Editing, ed. Ronald Gottesman and Scott Bennett (1970); and Medieval Manuscripts and Textual Criticism, ed. Christopher Kleinhenz (1976). During this time a collection of original essays on "Textual Studies in the Novel" appeared under the editorship of Warner Barnes and James T. Cox as a special issue of Studies in the Novel (7 [Fall 1975], 317 — 471).
Particularly Literary & Historical Editing, ed. George L. Vogt and John Bush Jones (1981); The Division of the Kingdoms: Shakespeare's Two Versions of "King Lear", ed. Gary Taylor and Michael Warren (1983); and Textual Criticism and Literary Interpretation, ed. Jerome J. McGann (1985).
In Germany and France it has increased, with the inauguration of Editio (1987 — ) and Genesis (1992 — ). Of new periodicals in English, I am aware only of New England Book and Text Studies (1994 — ), written by C. Deirdre Phelps.
Charles Rossman edited "A Special Issue on Editing Ulysses" for Studies in the Novel, 22 (Summer 1990), 113 — 269; and William P. Williams edited a "Special Issue on the New Oxford Shakespeare" for Analytical & Enumerative Bibliography, n.s. 4 (1990), 1 — 97.
The Davison volume (which celebrates the centenary of The Bibliographical Society in London) includes Sebastian P. Brock's "Developments in Editing Biblical Texts" (pp. 236 — 243), Fredson Bowers's "Notes on Theory and Practice in Editing Texts" (pp. 244 — 257), and John L. Flood and Conor Fahy's "Analytical and Textual Bibliography in Germany and Italy" (pp. 258 — 269), as well as my "Issues in Bibliographical Studies since 1942" (pp. 24 — 36), which contains a section dealing with textual questions. In the Barker volume, Barker himself provides an appendix surveying "Intentionality and Reception Theory" (pp. 195 — 201).
This volume is edited by Marianne Børch, Andreas Haarder, and Julia McGrew. Medieval texts were the subject of more entire anthologies in the early 1990s than were the texts of any other period. Others were The Editor and the Text, ed. Philip E. Bennett and Graham A. Runnalls (a 1990 festschrift for A. J. Holden), The Editing of Old English: Papers from the 1990 Manchester Conference, ed. D. G. Scragg and Paul E. Szarmach (1994), and the three named in the next three sentences of the text, plus the Faulhaber-Craddock collection commented on later. A common misunderstanding of recent editorial discussion is reflected in Szarmach's introduction to The Editing of Old English: "the dispute," he says, is "between those who see the possibility for the stable, the fixed, and the unchanging and those who see only process, continuing change, and varying relations of connections" (p. 2). A collection dealing with a different period is Victorian Authors and Their Works: Revision Motivations and Modes, ed. Judith Kennedy (1991).
In his fine review of this anthology — in Text, 6 (1994), 398 — 403 — David Yerkes understands that the equation of "textual" and "critical" is simply a manifestation of the role of human thought in constructing what we consider to be reality. "As Housman, following Aristotle, realized," he says, "to be alive is to want to know, even if 'knowing' cannot ever be anything more than mental exercise" (p. 399).
Machan, in "Middle English Text Production and Modern Textual Criticism" (pp. 1 — 18), expresses the need for "a more historically sensitive model" (p. 18) than intentionalist editors' concern (influenced by the New Criticism) for "a text to transcend time" p. 9). Hanna, in "Producing Manuscripts and Editions" (pp. 109 — 130), speaks of "the peculiar manner in which textual 'authority' is dispersed within medieval culture" (p. 120), resulting in a "plurality" of texts.
A third Toronto volume appeared during this period: Challenges, Projects, Texts: Canadian Editing, ed. John Lennox and Janet M. Patterson (1993).
Although I make no attempt below to cite the reviews that these anthologies have received, I do think it important to call attention to the substantial reviews of textual studies and editions that have been appearing in Text since its review section was inaugurated under Peter Shillingsburg in volume 6 (1994). That volume contains reviews of five of the anthologies taken up below (in addition to the review cited in note 36 above): the two 1990 anthologies, the Oliphant-Bradford and the Eggert, are reviewed by W. Speed Hill (pp. 370 — 382) and Tim William Machan (pp. 383 — 386), respectively; and three 1991 collections, edited by Cohen, Bornstein, and Barney, are reviewed by Michael Groden (pp. 366 — 369), Hugh Witemeyer (pp. 391 — 397), and Richard J. Finneran (pp. 387 — 390), respectively. The next volume of Text (volume 7, also dated 1994) contains D. C. Greetham's particularly thorough review (pp. 461 — 477) of another 1991 anthology, that edited by Small and Walsh.
This anthology was also made available in 1990 as a special number of the Library Chronicle of the University of Texas (20.1 / 2). The opening paragraph of Carver's introduction asserts that Bowers applied Greg's theory "in ways that seemed to the critic increasingly complex, even arcane" and thereby reinforced the split between the "establishment" and the "interpretation" of texts — a surprising claim in light of Bowers's insistence on the place of critical judgment in editing. Carver also repeats the canard that "the orthodox editorial theory of eclecticism developed under the sign of the New Criticism" (p. 8).
McGann's piece, "How to Read a Book" (pp. 13 — 37), was reprinted in his The Textual Condition (1991) and is mentioned in note 22 above; the concept of "radial reading," discussed in that essay, is commented on in my "Textual Criticism and Literary Sociology" (see note 2 above), pp. 137 — 138. McLeod's article is "from 'Tranceformations in the Text of Orlando Furioso'" (pp. 61 — 85), and McKenzie's is "Speech-Manuscript-Print" (pp. 87 — 109). Ian Willison's "Editorial Theory and Practice and the History of the Book" (pp. 111 — 125) — which contains the unperceptive remark that as an historian he is "obliged to favor" the socialized view of authorship (p. 113) — is briefly discussed (on the basis of an advance copy) in my "Textual Criticism and Literary Sociology" (see note 2 above), p. 130.
The titles of these three essays are as follows: Hellinga, "Editing Texts in the First Fifteen Years of Printing" (pp. 127 — 149); Warren, "The Theatricalization of Text: Beckett, Jonson, Shakespeare" (pp. 39 — 59); Gabler, "Textual Studies and Criticism" (pp. 151 — 165). Gabler's essay is also printed as the opening essay (pp. 1 — 17) of Editing in Australia, the anthology to be discussed in the next paragraph.
But to repeat: "historical existence" cannot be equated with what survives (or was once present) in physical form; and an eclectic text, though it may draw readings from documents widely separated in time, does not aim at synchrony but instead at the reconstruction of a given moment's intention (one of the successive "products" that make up a "process" — see my discussion of Shillingsburg below). (The title of the essay is "Textual Product or Textual Process: Procedures and Assumptions of Critical Editing" [pp. 19 — 40].) It is also inaccurate to suggest that critical editors have thought of their work as objective (p. 24); they have repeatedly noted that critical editions are products of judgment. (I certainly agree with Eggert's doubts about "the factualness of 'facts'" [p. 26]; see my "Printing History and Other History" [see note 15 above].) I do not wish to detract from the well-deserved praise Eggert accords to Peter Shillingsburg ("Shillingsburg's work has crystallised for a lot of people" an understanding that "the critical edition is indeed critical" [p. 34]) when I say that this point was being made in the 1960s.
The volume also includes — among others — a piece by Harold Love on "The Editing of Restoration Scriptorial Satire" (pp. 65 — 84), which should be read by all those interested in genealogical method (see also his 1993 book, Scribal Publication in Seventeenth-Century England); one by Stephanie Trigg ("Speaking with the Dead," pp. 137 — 149), on the "narrative subjectivity" (p. 139), or "editorial voice" (p. 148), in editions; and one by Jeff Doyle ("McLeoding the Issue: The Printshop and Heywood's Iron Ages," pp. 150 — 168), which draws on Randall McLeod's bibliographical work to show how "the process of typesetting is . . . a critical reading of the text" p. 165). (Gabler's is briefly discussed above as part of the Oliphant-Bradford collection, in which it also appears.)
Later in the volume, Jeff Doyle (see note 44 above) says, "There is no need for the process and production models to be separate" (p. 163). In his case, the point refers specifically to products (like printed editions of the Renaissance) that are not uniform but embody "the layerings of process" in jumbled form. (One could add that his point reinforces the need for historical reconstructions to supplement surviving documents.)
Another weakness of the essay (besides its unfortunate title) is its opening paragraph, which is full of questionable statements or implications. To take one of them: the familiar suggestion is made that a goal (such as "establishing what the author wrote") is undesirable because it is "beyond definitive reach"; a person might, Shillingsburg seems to imply, prefer a goal that is "perhaps more attainable." The attainability of a goal has nothing to do with the desirability of pursuing it; and most of the essay, I believe, attests to this point, for the idea of "definitive reach" is similar to that of "stability," and being "beyond definitive reach" is like instability in reflecting the nature of historical complexity. Shillingsburg's essay, in other words, is far more thoughtful and sophisticated than his introductory comments suggest. Occasional remarks scattered through the essay are also disappointing, as when he says that "one of the nice things about scorning historical reconstructions is that there is less work involved" (p. 49). This is an incomprehensible joke in the context, for Shillingsburg's comprehensive approach does not banish historical reconstructions, since it allows for emendations and an interest in intention.
The title, Cohen explains in his introduction (p. xvi), is derived from an essay of McGann's that speaks of "the angels of hermeneutics" avoiding the editors who "hurl defiance at the heavens of the interpreters." Those "devils," however, are the ones who have searched for "ideal" texts: Cohen continues the tradition of characterizing the dominant aim of Anglo-American editors as an "ideal text" (p. xi) and of linking their approach with the New Criticism — though the weakness of this association is actually revealed in the process (see the discussion on p. x). One of the reviews this book received — Howard Horwitz's in American Literature, 65 (1993), 198 — 200 — shows how extreme are some of the distorted views of the Greg-Bowers tradition: "This model of editing assumes that a work is an absolutely determinate and determinable object embodying an author's final intention."
Howard-Hill, "Variety in Editing and Reading: A Response to McGann and Shillingsburg," pp. 44 — 56.
Mailloux, "The Rhetorical Politics of Editing: A Response to Eggert, Greetham, and Cohen and Jackson," pp. 124 — 133; Greetham, "The Manifestation and Accommodation of Theory in Textual Editing," pp. 78 — 102; Cohen and Jackson, "Notes on Emerging Paradigms in Editorial Theory," pp. 103 — 123.
Cain, "Making Texts New: A Response to Gabler, McLaverty, and Grigely," pp. 195 — 203; McLaverty, "Issues of Identity and Utterance: An Intentionalist Response to 'Textual Instability,'" pp. 134 — 151; Gabler, "Unsought Encounters," pp. 152 — 166; Grigely, "The Textual Event," pp. 167 — 194.
The word "others" here represents a significant (and welcome) addition to the conventional concept of an apparatus; normally an apparatus records only "actual [i.e., documentary] utterances," but the scope is here enlarged to include the emendations that would need to be made in order to produce other intended utterances than the one(s) presented in the reading text(s).
Grigely correctly sees the need to distinguish "work" from "text," but there are two problems with his discussion. First, in defining a work (following McGann) as "an ongoing — and infinite — manifestation of textual appearances" (p. 176), he does not sufficiently account for intended texts as part of the sequential process. (He later recognizes performed as well as inscribed texts but still does not relate them to intended texts.) Otherwise the concept of a work as a succession of texts seems appropriate enough; but (and this is the second problem) Grigely does not follow his definition in his ensuing discussion, as when he calls the work of The Tempest "a Platonic form or idea" (p. 176). When, on the next page, he more formally defines a work as "a nontangible idea represented by a sequential series of texts," he has substantially departed from the earlier definition: to call a work an "idea" is very different from (and much less useful than) calling it a succession of texts.
Greetham chooses to pursue this interesting topic by a "rereading" of textual criticism "against the grain" of other "theoretical dispensations," specifically in this instance psychoanalytical criticism. This ingenious excursus does not, however, connect directly with what I would see as one of the basic aspects of the topic: how the "ideology" of an edition, as reflected in its apparent patterns of emphasis and subordination, relates to the editor's outlook toward other ideologies. An editor's choice of one approach does not necessarily imply a disapproval of others (and, indeed, an editor may say this explicitly) — a point that needs to be accounted for in any treatment of the "manifestation" of theory in editions.
Basic to the German approach, according to Gabler, is the idea of "authorization" that is "document-related" (p. 163). The editor "must present with historical faithfulness" a documentary text and yet can emend "indubitable textual errors" (p. 164). (I am assuming that the spelling "induitable" in this text is such an error and have corrected it in my quotation here.) The incoherence of this approach has been repeatedly pointed out in connection with the "best-text" editing of medieval writings: if one is to undertake critical editing, one cannot be selective in applying critical thinking to a documentary text. Cf. my review of Gabler's edition of A Portrait of the Artist as a Young Man in Common Knowledge, 3.3 (Winter 1994), 164 — 169.
It does contain, to my view, occasional lapses, however. For example, there is the implication (pp. 270 — 271) that author-centered theories are more vulnerable to logical criticism than socialized theories because they place the literary work in the author's mind. But the real distinction here is between approaches that focus on documentary texts and those that focus on works; the latter have to posit the mind as the source of authority, but it need not be an author's mind (for authority could also be seen as residing in the minds of a collaborative group or in a reader's mind). Another problem is raised by the statement that "Psychoanalysis, Marxism, and feminism offer three theoretical positions from which human agency and autonomy are strenuously questioned" (p. 276). The implication that intentionalist editing presupposes greater human autonomy than would be consistent with these theories is a misconception; a writer's intention to place certain words in a text is not seen by intentionalist editors as totally isolated from life.
Walsh, "Bentley Our Contemporary; or, Editors, Ancient and Modern," pp. 157 — 185; Small, "The Editor as Annotator as Ideal Reader," pp. 186 — 209. Walsh's essay ends with an admonition not "to privilege the critic's quest for significance to himself over the interpreter's quest for author's meaning"; Small's ends by stating that "a concept of authorial intention . . . is more useful — in the sense that it does more work — than any other theory." That these are well-considered conclusions is evident from the knowledgeable discussions that precede them. (The other essays in the volume deal with particular authors but frequently engage theoretical issues sufficiently to make them of interest to scholars in different areas.)
Most university press directors would be surprised to hear that "their presses may in the act of constructing and disseminating authoritative texts also legitimate themselves as the places where such texts are constructed and disseminated" (p. 2).
Ian Small, in his review of this anthology in English Literature in Transition, 38 (1995), 195 — 203, takes the occasion of Shillingsburg's essay to note that "there is a logical contradiction waiting to ambush the postmodernist editor": if "'works' are constructed by the values and prejudices of the editor" and "'versions' are simply there to be revealed," then there is "no necessary relationship between versions and works," and thus "it is difficult to see why we need to know about versions" and indeed why we should "bother with editions of any kind." He proceeds to make the important point, frequently overlooked, that versions "are not simply 'there' to be revealed" but, like works, "have to be identified" — in a process that "willy-nilly involves values and prejudices."
The essays referred to in this paragraph are Greetham's "Editorial and Critical Theory: From Modernism to Postmodernism" (pp. 9 — 28), Shillingsburg's "Polymorphic, Polysemic, Protean, Reliable, Electronic Texts" (pp. 29 — 43), and Williams's "I Shall Be Spoken: Textual Boundaries, Authors, and Intent" (pp. 45 — 66). These essays are drawn from the first section of the volume, entitled "Editorial Theory Today"; Parts 2 and 3, with more specialized essays, are entitled "Editing Literature" and "Editing in Other Disciplines."
Hill's essay is a revised version of his contribution to Greetham's Textual Scholarship (see note 6 above), which had not been published at the time. As for the RETS programs of the years mentioned in the subtitle, the one for 1987 is represented by a single lecture instead of the usual program of three or four papers; but the papers from the other years were originally distributed to the Society's members in the form of annual pamphlets made up of photocopies. One of them has been commented on in my "Textual Criticism and Literary Sociology" (see note 2 above), footnote 63.
The either/or approach emerges at many points, as when Hill says that "genealogical editions betray their origins in early nineteenth-century romantic ideology of the autonomous author/artist whose work derives its value from its unique authorial origins, not its subsequent social distribution" (p. 8). What this sentence betrays (through the "not") is both a belief that these interests cannot be held simultaneously and an unfair attribution of the same rigidity to editors who have produced genealogical editions.
This collection, entitled "Textual Scholarship and American Literature," was published as a special issue (20.2) of Resources for American Literary Study. Cohen indicates, in a note to his introduction ("Textual Instability, Literary Studies, and Recent Developments in Textual Scholarship," pp. 133 — 148), that this collection will be incorporated, along with additional essays, into an anthology to be entitled Texts and Textuality.
His essay, entitled "'Why Don't They Leave It Alone?': Speculations on the Authority of the Audience in Editorial Theory" (pp. 85 — 99), gets off to a misleading start, for the opening paragraph notes a recent shift "away from the hidebound assumptions and defensive postures that had characterized a small but entrenched Anglo-American editorial establishment"; but, with this gesture out of the way, the essay builds on Eaves's obvious understanding that his position can stand on its own. The same even-handedness that Eaves displays is also apparent in the editors' introduction, where they say, "The material artifact is always resistant to critical appropriation; however, this does not mean that the act of seeking a glimpse of the historical past is without value" (p. 3). Not all the essays in the volume deal with inscribed texts, but some that do are contributed by J. Paul Hunter (on Sterne and Pope), Hamlin Hill (on Mark Twain), Jerome J. McGann (on Pound and other "modern and postmodern poetries"), and Jeanne Holland (on Dickinson). The essays were originally written for a conference on "Textual Technologies: Text, Image, and History" at Texas A & M University on 26 — 29 March 1992.
The one on German work, to be part of the Michigan series, is Contemporary German Editorial Theory, ed. Hans Walter Gabler, George Bornstein, and Gillian Borland Pierce; the one on the French approach is the special issue of Romanic Review mentioned in note 20 above; and the Greetham volume is described in note 6 above.
The same general argument would hold, of course, for an interest in any kind of intention, whether authorial or not. Intended versions (regardless of whose intention is being focused on) cannot be equated with documentary texts and therefore must be critically constructed by an eclectic process. A more recent article on the Lear question also concludes that a "refusal to countenance conflation, with reference especially to F's major omissions, is to risk a blinkered and naive reinforcement of changes that may originally have been made to the play for very questionable reasons" (such as the embellishment of Edgar's role for a star performer); see Robert Clare, "'Who is it that can tell me who I am?': The Theory of Authorial Revision between the Quarto and Folio Texts of King Lear," Library, 6th ser., 17 (1995), 34 — 59 (quotation from p. 59). Another article (though not on Lear) that defends eclecticism is James L. W. West III, "Fair Copy, Authorial Intention, and 'Versioning,'" Text, 6 (1994), 81 — 89: he believes that editing versions is feasible for short works but not for long ones, and thus for the latter it is necessary to "apply one's critical intelligence to the surviving drafts, with or without an existing fair copy, and attempt to create an eclectic ideal" (p. 88). Leaving aside the inappropriateness of the word "ideal," this statement does not appear to acknowledge that even a single eclectic text represents a version (a particular moment in the history of a work) and that multiple critical texts of different versions would all be, in principle, just as eclectic as that of the (presumably) final authorial version suggested by West. It is a sign of the insidiousness of the tendency to equate versions with documentary texts that even West's article seems to have a trace of it. He distinguishes situations where there is an authorial fair copy from those where there is not; yet operationally there is no difference, since in either case (as West knows) one must use "critical intelligence" and not simply accept a single documentary text. And when he says that the idea of presenting versions assumes "that each embodiment of the text chosen for reproduction possess some measure of finality" (p. 87), there is a hint that the "embodiments" are the versions, whereas versions actually require critical (eclectic) activity for their construction. One of the many examples of failure to accommodate this point is Grace Ioppolo's Revising Shakespeare (1991), in which the chapter on "Revising King Lear and Revising 'Theory'" (pp. 161 — 187) treats as "apparent" the idea that "the conflated physical text is fraudulent because it synthesizes and reduces the multiple versions produced by the author in the process of revision" (p. 162). She asserts unequivocally, "Any edition of King Lear which conflates the Quarto and Folio texts . . . produces . . . a counterfeit and non-Shakespearian foundation upon which only the most limited literary interpretation and meaning can be built" (p. 181). Like a number of the revisionists, she gives the impression that editors who engage in conflation are motivated by the desire to maintain an image of Shakespeare as a writer who never blotted a line; but whether or not he revised is a separate question from whether conflation is necessary to produce what he intended at any given time. Another egregious instance of the problems created by the equation of versions and documentary texts is the section of Jack Stillinger's Coleridge and Textual Instability (1994; see note 13 above) entitled "A Practical Theory of Versions" (pp. 118 — 140). In the "five-point scheme" (p. 132) of his "practical theory," the first point is that "A version of a work is a physically embodied text of the work"; yet the third point is that it is an arbitrary matter to determine how much difference between texts is necessary in order to regard them as separate versions. The two points are at odds: if versions are the texts of documents, then obviously different documents contain different versions. His "practical" approach means that he has "cheerfully ignored" the fact that texts are "physically embodied," believing that words and punctuation can readily be transferred to other embodiments — even though he is thereby ignoring the "knottiest problem in textual theory" (p. 133). Whether or not one wishes to call this a "theory," one can hardly call it "practical," since it is not carefully thought through and is therefore not usable.
Especially his Scholarly Editing in the Computer Age (1984; revised, 1986). I have commented on this book in SB, 39 (1986), 39 — 45 [147 — 153]. Another related article of his that appeared in the same year as the one to be discussed below (and can be thought of as a pendant to it) is "Textual Variants, Performance Variants, and the Concept of Work," Bibliographical Society of Australia and New Zealand Bulletin, 15 (1991), 60 — 72 (reprinted in Editio, 7 [1993], 221 — 234). He has also, more recently, provided a useful historical sketch and summary of the current theoretical debates, reaching the sensible conclusion that the "anxiety of text is a good and normal anxiety," which is "best relieved, but never cured, by open investigation by critical minds"; see "Textual Angst," LiNQ [Literature in North Queensland], 21 (1994), 71 — 93.
I have not quoted the definition of "version" from Shillingsburg's appendix of definitions (which provides the definition of "work" quoted in the first half of the preceding sentence) because I do not find the definition there to be stated in a way that is consistent with the discussion in the body of the article. In the appendix, "version" is formally defined as "a concept by which Material Texts . . . are classified as representative" of what he calls "Potential," "Developing," or "Essayed" versions; although the intangibility of versions is recognized here to some extent, this definition does not seem to me to make sufficiently clear that any material text might reflect more than one version and that it is not the material texts as entities that are being classified. Indeed, the definition should, I think, be parallel to that of "work," in order to show the parallelism in the concepts and to make clear that versions are subsets of works.
I find some of Shillingsburg's explanations of his terms puzzling, however — indeed, so much so as to constitute a serious weakness in his presentation. For example, he says, "The Essayed Conceptual Text is always manifested in a physical form, but it is not a physical or Material Text, for the Conceptual Text that is Essayed remains (as the author's mental concept) invisible and probably not stable; but the embodiment of the Conceptual Text is visible and fixed in a material medium" (p. 52). If Shillingsburg had written "often" instead of "always" as the sixth word of this sentence, I would see no problem; but as it is, he seems to be claiming that all finished texts in authors' minds get written down or otherwise placed in physical form. Clearly this is not true. Whether or not there is any point in trying to recover a mental text that was not (however imperfectly) "embodied" is a separate matter, and one that should not affect the definition of an "Essayed Conceptual Text." If the definitions of these terms are to promote rigorous thinking, they must cover all theoretical possibilities. The same problem reappears in the appendix of definitions, where "Essayed" is defined as "finished (at least temporarily) versions as evidenced by completed manuscripts or revised texts" (p. 81). But the existence of such versions does not depend on their being "evidenced" in this way. Furthermore, such physical evidence as there is may be a mixture of versions, and Shillingsburg's definition does not guard against the inference that documents preserve versions (something he is generally careful to avoid). (There is further laxity here: the plural "versions" falls under the singular heading "Version," and "manuscripts" is used to mean "manuscript texts," obviously an important distinction in this context.) The related definition of "Conceptual Text" is "Any text that is 'held' in the mind or contemplated by a person. Conceptual texts are the only kind that can be experienced, though Material Texts are where they begin." What Shillingsburg has done here is to limit the definition to the readers' perspective: texts in readers' minds "begin" with the physical texts they read. But if the definition is to cover authors as well (as it should, since "person" includes all persons), it is wrong to say that mental texts "begin" in physical form. A similar tangle lies behind the ambiguous definition of "Linguistic Text": "A Sign Sequence for an Essayed Version displayed in a Document." If the documentary display is essential to the existence of a "Linguistic Text," then the definition would seem to be at odds with that of "Material Text": "The union of Linguistic Text and Document: A Sign Sequence held in a medium of display" — where the "Linguistic Text" is not in itself physical. Yet in the body of the essay we are told (p. 53) that "if there is no Material Text there is no Linguistic Text"; and just before that (p. 52), a "Linguistic Text" is said to exist "first as a Conceptual Text (thought) then as a Semiotic Text (sign), and then as a Material Text (paper and ink or some other physical inscription or production)." But it is not clear why a "Linguistic Text" must be physical if a "Semiotic Text" can be mental. Of course, Shillingsburg can define "Linguistic Text" this way if he wishes, but he seems to have defined it in two different ways. Such elementary blunders are disturbing, especially in a piece that seems in many respects to have been carefully thought out.
The full statement is that textual criticism is "the science or art of detecting and removing textual error, the discipline of establishing what the author wrote or final authorial intention, the work of purifying and preserving our cultural heritage." This definition actually limits itself to editing, describing the activities that editors have generally (before the past half-century) thought of themselves as engaging in; but "textual criticism" has usually been regarded as a much broader field — the critical examination of textual histories, whether or not an edited text was the result. Thus one could disapprove of, or not be interested in, the activities enumerated in Shillingsburg's definition and still respect, or have an interest in, the field of textual criticism.
In the passage that follows this statement, Shillingsburg pursues — mistakenly, I believe — an analogy with music. Whereas the reader of literature both constructs the "Reception Text" and engages in its interpretation, in the case of music, he says, the orchestra constructs the "Reception Text" and the listener responds to it. But this view oversimplifies the situation. To use Shillingsburg's terms, what the orchestra creates is indeed a "Reception Text," as far as the orchestra is concerned, for that text is the orchestra's interpretation of the printed score. But from the audience's point of view, the orchestra has engaged in a "Production Performance," providing a text to be responded to; what the audience makes of it is then a full-fledged "Reception Performance," involving the construction of both a "Reception Text" and a response.
Paul Morgan — in "Text and Authenticity: Examining the Terminology," Bibliographical Society of Australia and New Zealand Bulletin, 16 (1992), 141 — 147 — says that Shillingsburg's essay "certainly seems the right direction to take" and suggests the establishment of a standards committee to formulate "a common bibliographic terminology" (p. 147). Another attempt — very brief, in contrast to Shillingsburg's — to define basic terms is Rolf E. Du Rietz's "'Work,' 'Text,' and 'Document' as Bibliographical Concepts: A Short Note," in Mercurius in Trivio: Studi di Bibliografia e di Biblioteconomia per Alfredo Serrai, ed. Maria Cochetti (1993), pp. 65 — 70. Although he does not, unfortunately, deal with the relation of versions to works, he does tackle the immateriality of sequential works by distinguishing (on p. 66) an "ideal text" ("the sequence in a sequential work") from a "natural text" ("the sequence resulting from any attempt to record, reproduce, reconstruct, perform, or communicate a sequential work"). The latter, however, is still, in his view, an abstraction, since the sequence can exist apart from a physical embodiment; and he therefore feels that he must further distinguish (on p. 68) the "text of the document" (the "natural text," a sequence) from the "documentary text" (the "part of a document which consists of the 'material' text"). Since he has previously said that inked letterforms are not part of the text, but rather are part of the document, it is not clear what the "material text" consists of. One understands that he wishes to exclude typography as an intended element of the work; but since it is used to record the intangible sequence of words, I do not see how one can define the "material" text without reference to it. (His use of "ideal" seems to me to raise another problem, since he believes the fact that ideal texts "may be known to us only indirectly" makes them "like Plato's ideas.")
Eggert elaborates his thoughtful criticism of McGann in a "Coda" appended to the essay ("A Commentary on The Textual Condition by Jerome McGann," pp. 17 — 24). The heart of the criticism is that McGann, in his emphasis on texts as social events, neglects the role of individuals within that process (and the research that would lead to the determination of individual responsibility). "I fail to see," Eggert says, how McGann can insist on "the necessity of retrieving what he repeatedly calls the 'determinate circumstances'" of textual production, "while edging away from, or blurring with generalities, the most focused form of it — individual agency" (p. 23).
I concentrate on this element in Eggert's article because it is the most basic one. The question that he calls the "subject" of the last third of the essay — whether "textual bibliography" "can or should alter its methods to deal with this expanded conception of the work" (p. 10) — is a non-issue. Given the usual definition of "bibliography" to mean the study of physical evidence, it is perhaps not surprising that he should consider "textual bibliography" (if this unfortunate term can really be thought to designate a field that exists) to be "best equipped to deal with the work at the level of documentary inscription" (p. 11). But the evidence of the continuing life of a work and of readers' responses (constituting the "expanded conception of the work") would normally be in physical form also, and bibliography could handle it just as well as it can deal with initial inscription. But textual scholars have to do whatever work is necessary in any case, and it does not matter whether or not the work is called "bibliography."
I am not suggesting that Eggert is unaware of this point. He comes close to it in a footnote at the end of his "Coda," where he says, "The persistence of a language held largely in common allows a reading of the document, but the different contexts of writing and reading mean inevitably that the two 'readings' (writer's and reader's) will differ" (p. 24). Even here the act of recognizing a language is not clearly differentiated from the "readings" that follow, and I do not find that he explores the implications of this distinction for his approach.
I have tried to emphasize the place where — given only two terms — Eggert has chosen to put the dividing line between them, rather than to criticize the terms themselves, since there can be no "right" way to define the particular words "text" and "document." But I must add that his usage, if adopted, would surely produce further confusion, since "document" is so regularly used to refer to an entire physical object that limiting it to one part of an object, a physical text, forces readers constantly to make a mental adjustment when they come across the word. Besides, all the other features of the object need to be covered by some term; although Eggert is not concerned with them here, they figure prominently in many essays in textual theory, since (as Eggert understands) all of them can play a role in readers' responses. For these reasons, I have preferred to use the phrase "text of a document" (or "linguistic text of a document," when required by the context) to refer to the part of an object that Eggert calls simply "document," so that the word "document" is still available for the object as a whole. Then I would say "text of a work [or version]" to refer to the mental construct in an author's or reader's mind (rather than simply "text"), showing through the parallelism of the phrases that a linguistic sequence can exist both tangibly and intangibly. Not everyone will find these phrases satisfying, of course, but Eggert's terms are less satisfactory. The imprecision of his terms is shown several times in his own essay. In one sentence, for example, he says "material object — or what I prefer to call document" (p. 2), as if "document" stands for the whole object; but in the next sentence the word "document" is followed by the footnote I mentioned earlier, in which he says, "For the purposes of the present discussion I am primarily interested in the document considered as the physical inscriptions it bears." The word "primarily" adds another confusion, since the next sentence begins, without qualification, "I leave aside the other kinds of meaning encoded in the physical nature of the document." And that sentence ends by saying that — when one does wish to pay attention to these other physical features — "the document itself must be 'read' textually" (where "document itself" is the whole object, as again in his footnote 21). Furthermore, to speak of reading physical features "textually" is awkward in light of the special meaning given to "text" (which is itself awkwardly equated on the same page with "textual meaning").
And he rarely engages in reductive dismissals of earlier work. But an exception is this sentence near the end: "The sort of stability editorial theory aspires to is thus undercut by its own activities: it seeks to maintain an author-centered status quo, whereas the author-centered moment never existed as a real event, nor can it" (p. 58). This sentence may set a record for the density of its distortions. But to take up only the most consequential one: the moment when an author is producing a text that seems — at that moment — to be finished is certainly an "author-centered moment" (regardless of the influences on the author's thought) and is certainly "a real event."
His definition of textual criticism early in the essay is an appealing proleptic glimpse of his final vision: "a means and a process by which careful observation of textual variations and textual contexts will lead us towards a better understanding of why those variations should exist and what they might mean both for ourselves and for others" (p. 32). But "textual variations and textual contexts" implies a more delimited sense of "text" than is ultimately offered.
This formulation of course accommodates instability, since all objects change over time. As Heller notes, it is "only by convention" that we can think of whole stable objects at all (p. 22); an object is not "an enduring spatial hunk of matter" but "a spatiotemporal hunk of matter," the parts of which are constantly changing (p. 4).
This division is effectively undercut by the word "usually" in another of his sentences: the meaning of a text, he says, "is not usually got by any act of interpretation" (p. 339). His justification for "usually" is the situation in which one employs "contextual clues" to discover the meaning of a text in an unknown language. But there is slippage here in what "meaning" signifies; this use of context to determine textual meaning is not unlike the resolution of ambiguity, which Currie places within the interpretation of works. His whole analysis would have been helped by a concept of "text of the work," which would allow him to show the role of meaning in constituting wording at the level of work. Early in the article he says that an author's intention to spell correctly leads us to correct the text (p. 326), and this would have been a natural place to introduce the idea of "text of the work," but he does not do so, leaving open the question of the status of the corrected text (and the relation of "texts" to tangible objects).
For example, in his discussion of literary works on pages 167 — 171, he does recognize that authors sometimes pay attention to the details of physical production and that readers are not always interested in authorial intention; but he does not develop either point or explore what role physical features can play in studying a literary work. When he says that "one copy of a novel provides as good an access to the artifact/art work as any other" (p. 167), he is not making quite so incautious a statement as one might at first think, if one takes the context into account; what he means is that if the text is as the author intended it to be, then one copy is as good as another. Even so, all the statement accomplishes is to say that literary works are intangible entities (or, in his terms, "abstract artifacts"), and the ensuing discussion does not proceed very far in examining the relation of the "idealized artifact" to the "physical exemplar" (p. 170).
This is not the only problem with Gracia's distinction between "text" and "work." He defines "work" as "the meaning of certain texts for which society has developed rules so they fulfill a specific cultural function that renders them works" (p. 68). In defining works as meanings, he does not always avoid the danger of undervaluing the role of form in contributing to the meaning of verbal works. Although he cites the rules for the sonnet form as an example of the socially determined rules that are essential to his definition of "work" ("The meaning of a sonnet is a work because the text adheres to certain rules that are supposed to apply to sonnets" [p. 67]), he then claims that such formal considerations do not apply to "works devoid of artistic and literary quality" (p. 69). He is therefore reduced to claiming that "textual works are the meanings of texts, except in cases where the texts have an artistic dimension, for then the works are the meanings plus whatever elements are essential for the identity of the works in question" (p. 69) — a statement that untenably separates meaning from form and presupposes a firm distinction between literary and nonliterary works. (This may be an appropriate place to call attention to a pair of essays that — although written by a lecturer in English — deal with philosophical concepts: John Winter's "Textual Criticism and Ethics: An Inquiry" and "The Application of Ethics to Textual Criticism," Bibliographical Society of Australia and New Zealand Bulletin, 18 [1994], 187 — 198, and 19 [1995], 31 — 46. Winter's aim, as stated in the second piece, is "to recognize the human qualities of the conditions in which textual critics deal" [p. 42].)
Though it is not without its share of questionable statements, such as the anachronistic idea that the "New Bibliography" was named in reaction to the "New Criticism" (p. 2), or the notion that until recently the established bibliographical journals were inhospitable to theory (p. 9), or the suggestion that Gabler's edition of Ulysses is important because "for the first time it subjected a major work of literature written in English to the sort of genetic examination that had hitherto been primarily used on European texts" (p. 16) — a claim that can hardly be defended in view of Hayford and Sealts's 1962 edition of Billy Budd, Sailor (to name an edition with a running genetic text; but one must remember that texts presented in other ways can still reflect "genetic examination").
These contributions, by W. Speed Hill, Mary-Jo Kline, Joel Myerson, David J. Nordloh, and Donald H. Reiman, are gathered under the heading "The Politics of Editing" in Text, 6 (1994), 91 — 132. Myerson's piece ("Editing and Politics") is a revised version of his 1990 presidential address to the Association for Documentary Editing, published as "The Politics of Editing" in Documentary Editing, 13 (1991), 1 — 3. Another paper in this genre delivered at an ADE meeting is W. Speed Hill's "The Editor on the Iceberg: or, Just How Far South Is the Gulf Stream?", Documentary Editing, 12 (1990), 18 — 21. And Gary Taylor spoke on "The Rhetorics of Reaction" at the 1988 Toronto conference (his extended paper was not published until 1994, in Randall McLeod's anthology Crisis in Editing, pp. 19 — 59), where he described textual criticism as "an apology for, or a prolegomenon to, the exercise of power": "Editing exercises power, and it can only be understood by an analysis of power" (p. 19). (Some of the recent detailed studies in the history of editing deal more effectively with the cultural and ideological contexts of specific editions; see, for example, Peter Seary's Lewis Theobald and the Editing of Shakespeare [1990] and Neil Fraistat's "Illegitimate Shelley: Radical Piracy and the Textual Edition as Cultural Performance," PMLA, 109 [1994], 409 — 423.)
For some background on the German and French developments, their relation to textual criticism in English, and the few pre-1990 discussions in English, see my "Textual Criticism and Literary Sociology" (see note 1 above), pp. 112 — 118. There has been a great deal of activity in textual study in Germany and France during the early 1990s but little discussion of it in English (aside from the forthcoming anthology on German work mentioned in note 64 above). The fullest treatment of the French school is Almuth Grésillon's Éléments de critique génétique: Lire les manuscrits modernes (1994). In addition to the German and French periodicals Editio: Internationales Jahrbuch für Editionswissenschaft (1987 — ) and Genesis: Revue internationale de critique génétique (1992 — ), there have been several anthologies in recent years: Sur la génétique textuelle, ed. D. G. Bevan and P. M. Wetherill (1990); L'Écriture et ses doubles; genèse et variation textuelle, ed. Daniel Ferrer and Jean-Louis Lebrave (1991); Zu Werk und Text: Beiträge zur Textologie, ed. Siegfried Scheibe and Christel Laufer (1991); and Les manuscrits des écrivains, ed. Louis Hay (1993). (Hay was responsible for several earlier anthologies: Avant-texte, texte, aprèstexte [edited with Péter Nagy, 1982], La manuscrit inachevé: écriture, création, communication [1986], La naissance du texte [1989].)
Such as Jo Ann Boydston's address quoted in the next paragraph (from Text, 5 [1991], 1 — 13); Clayton J. Delery's "The Subject Presumed to Know: Implied Authority and Editorial Apparatus," Text, 5 (1991), 63 — 80; and Fredson Bowers's "Why Apparatus?", Text, 6 (1994), 11 — 19. (The insistent nature of the subject of apparatus is suggested by the fact that it was on Bowers's mind to the end; his article in the 1994 Text was a paper he had been scheduled to read before the Society for Textual Scholarship on 11 April 1991, the day he died.) Annotation is the subject of such articles as Ann Middleton, "Life in the Margins, or, What's an Annotator to Do?", in New Directions in Textual Studies (see note 40 above), pp. 167 — 183; Patrick S. White, "Black and White and Read All Over: A Meditation on Footnotes," Text, 5 (1991), 81 — 90; Richard Knowles, "Variorum Commentary," Text, 6 (1994), 35 — 47; Ronald Schuchard, "Yeats's Letters, Eliot's Lectures: Toward a New Focus on Annotation," Text, 6 (1994), 287 — 306; and James Woolley, "Annotation: Some Guiding Considerations," East-Central Intelligencer [East-Central/American Society for Eighteenth-Century Studies], n.s. 8.1 (January 1994), 11 — 16 (as well as Stephen A. Barney's 1991 anthology, Annotation and Its Texts, commented on above). An example of a theoretical article that includes a discussion of apparatus is D. C. Greetham's "Editorial and Critical Theory: From Modernism to Postmodernism" (see note 59 above), which attempts to defend, in its opening pages, the notion that forms of apparatus are implied by theoretical positions. (For a somewhat different approach, as well as a fuller discussion of apparatus than I have included in the present article, see my "Textual Criticism and Literary Sociology" [see note 2 above], pp. 133 — 140.) Greetham's discussion of the "postmodernist advantages" of "computer hypertext and its myriad permutations" (p. 16) does not keep in focus the distinction between using editions for historical study and using them for nonhistorical reading.
This article, from Romance Philology, 45 (1991), 123 — 148, is part of the special issue on the textual criticism of medieval literature that is briefly discussed as an anthology above. Jerome McGann has been an active advocate of hypertext — indeed hypermedia — editions. For a recent outline of his approach, see his "The Complete Writings and Pictures of Dante Gabriel Rosetti: A Hypermedia Research Archive," Text, 7 (1994), 95 — 105. Many of his discussions of the implications of hypermedia have also appeared on the Internet. See also Hypermedia and Literary Studies, ed. Paul Delany and George P. Landow (1991).
Including every variant page within an edition, resulting from stop-press alteration or the cancellation and substitution of leaves or gatherings.
Although linking is an enormous advance over what had previously been available for locating variants and assessing them in context, there is still a need in electronic editions for lists of variants. The ability immediately to locate a variant in a different text from the one being read at a given moment does not obviate a record that provides an overview of all the differences in all the texts (which can of course be usefully subdivided into categories — that is, coded so that the variants can be retrieved according to various classifications).
These ideas about electronic editions — along with the point that genetic study, which regularly limits itself to documentary texts, is in fact usually interested in intended versions — are set forth more fully in my "Critical Editions, Hypertexts, and Genetic Criticism" (see note 20 above).
The wording of this sentence rests on my definition of "primary record": "a physical object produced or used at the past moment that is the subject of one's inquiry." This definition appears in my "The Future of Primary Records," forthcoming in volume 58 (1996) of the Encyclopedia of Library and Information Science, ed. Allen Kent. It also underlies the "Statement on the Significance of Primary Records" adopted by the Modern Language Association of America and forthcoming in Profession 95.
All three of these facsimiles have recently been discussed in detailed examinations of the nature of facsimiles: Hinman's and Warren's in Joseph A. Dane's "'Ideal Copy' versus 'Ideal Texts': The Application of Bibliographical Description to Facsimiles," Papers of the Bibliographical Society of Canada, 33.1 (1995), 31 — 50; Vander Meulen's in James McLaverty's "Facsimiles and the Bibliographer: Pope's Dunciad," Review, 15 (1993), 1 — 15. Dane's piece, as its title suggests, is particularly relevant here, since it links the idea of eclectic editing (said by some to result in "ideal texts") with the production of composite facsimiles (said by some to result in "ideal copies") and thus raises the whole issue of the historical value of eclecticism.
In "Editing without a Copy-Text," SB, 47 (1994), 1 — 22, I tried to outline an approach for focusing on whatever set or sets of circumstances one prefers when preparing critical texts. Its rationale was offered as a replacement for Greg's, in two senses: the essay pushes editorial reliance on judgment to its logical conclusion (which Greg — and, indeed, Bowers as well — did not quite reach) by dispensing with the idea of a "copy-text"; and the approach is equally applicable to any editorial emphasis, not simply that of final authorial intention.
| | |