SOURCE: “Is There a Romantic Ideology? Some Thoughts on Schleiermacher's Hermeneutic and Textual Criticism,” in Text: Transactions of the Society for Textual Scholarship, edited by D. C. Greetham and W. Speed Hill, AMS Press, 1988, pp. 59-77.

[In the following essay, Rajan provides an overview of the development of Schleiermacher's hermeneutic philosophy in the context of Romantic ideas regarding discourse, criticism, and creation.]

Since the appearance of Jerome J. McGann's A Critique of Modern Textual Criticism in 1983, we have begun to reconceive significantly a textual criticism based on a theory of final authorial intentions. In taking issue with traditional textual criticism, McGann attacks both the notion of the author as an autonomous and unitary subject and the related notion of a definitive text that reflects the author's final intention and protects it from the errors of transmission and the intrusion of other voices. My argument is not with McGann's critique, but with his association of modern textual criticism with what he calls the “Romantic ideology.” The emphasis on “the autonomy of the isolated author” to the exclusion of any recognition that the mode of existence of the text is also social is, he says, grounded in “a Romantic conception of literary production.” And again, G. Thomas Tanselle's idea that the critical editor must seek to publish the text “which most nearly reflects the author's autonomously generated text,” even if such a text is hypothetical, McGann argues is founded on “a Romantic ideology of the relations between the author, his works, his institutional affiliations, and his audience.”¹ Writing in a similar vein, Lee Patterson sees Romantic hermeneutics as an idealist repression of what he calls “the analytic dissolutions of historicism,” the unwilling deconstruction of transcendent authors and ideal texts brought about by Wolf and Eichhorn and resisted by the Romantic period's alleged separation of theoretical hermeneutics from the material study of written documents.² My own contention is that not only is the ideology of the author as sovereign subject that subtends much textual criticism a modern reconstruction that finds only partial authority in the Romantic period but that the origins of the contemporary questioning of this authority initiated by McGann are also to be found in the Romantic period.

Romantic habits of composition, for one thing, belie any unitary theory of intention. The editorial work done on the Cornell Wordsworth shows that Wordsworth's poems went through various visions and revisions and that in many cases the moment of definition constituted by official publication was deferred through private circulation of manuscripts or repeated through the recontextualising of poems in different arrangements or the absorption of autonomous pieces into longer works.³ Such revisions, particularly within the dialogical context of circulation, reveal how the authorial voice is articulated on the trace of other social and literary voices. Wordsworth's reluctance to publish is apparent in an 1842 letter to Aubrey de Vere: “Publication was ever to me most irksome; so that if I had been rich, I question whether I should ever have published at all.”⁴ Although pressure from Coleridge sometimes led Wordsworth to see his poems as forming a canon, with The Prelude as antechapel to an incomplete cathedral, there is also evidence that he had grave reservations about monumental structures: that he wrote what Coleridge dismissively calls a “multitude of small poems” as a way of putting off entombment in the “great philosophic poem” and that he composed even his longer poems in discrete blocks which he had then to place within a confining connective structure.⁵

But a reluctance to claim finality and authority—though all the more striking in a period that seems to exalt the self—is surely not unique to the Romantic writer. Equally important, in the late eighteenth and early nineteenth centuries, is that we also witness the beginnings of a scholarly critique of the transcendent author and of editing as a means of unmediated access to the text intended by this author. This critique begins, as is well known, with the work of F. A. Wolf and J. G. Eichhorn on the gospels and Homeric poems. Although Wolf speaks initially of restoring the Homeric texts “more nearly to their ancient and original form,” he ends by conceding that this original form cannot be “laid out save in our minds” and that there may not even have been a Homer.⁶ His careful research disintegrates the notion of the Homeric poems as created by a single intention, replacing it with the idea of a historically collaborative authorship that has produced a composite text made up of layers of altering, editing, and imperfectly disguised attempts at unification. Moreover, an awareness of authorship and transmission as problematic constructs is not confined to classical and biblical scholarship. The roughly contemporary editing of the ballads raises similar questions about the recovery of original texts and the existence of single authors—questions that cannot be answered by mystical concepts of a folk authorship that is collective but somehow also unitary. Problems of composite authorship similarly surface in the editing of Beowulf, first catalogued in 1705, but not systematically edited until the nineteenth century.⁷

The work of the earliest textual scholars, however, is not theoretically radical. G. J. Thorkelin, who edited Beowulf in 1807 and 1815, and N. F. S. Grundtvig, who translated it in 1820, held to the idea of a single authorship. It was not until 1840 that Ludwig Ettmüller suggested that the poem might have been put together from separate lays and further emended by clerical editors, and not until 1869 that K. Müllenhoff undertook a systematic analysis of Beowulf in such terms. Because classical scholarship had a longer history, it arrived earlier at empirical findings that unsettled textual authority. But again, Wolf's own assumptions were hardly radical, being generated by the gap between his theoretical desire to privilege the text produced by the author and his material finding that the Homeric texts were compositely authored. In other words, because he remained committed to the idea of an original text, he produced a stemmatic theory that arranged textual variants or layers in an order of descent away from a lost origin. This paper, therefore, takes as its starting point a slightly later critic, Friedrich Schleiermacher, and specifically his work on the Gospel of Luke. On a descriptive level Schleiermacher is very much in the debt of scholars like Wolf and Eichhorn. But it is on a theoretical level that we must approach him if we are to understand why Coleridge saw him as the polar opposite of Eichhorn.⁸

My focus on Schleiermacher as a way of unsettling the Romantic ideology may seem puzzling, since a long line of theorists, beginning with Dilthey and including Gadamer and Ricoeur, have associated him with a psychologism that largely bypasses textual criticism.⁹ Since the appearance of Dilthey's influential essay “The Rise of Hermeneutics,” Schleiermacher has been credited with a distinction between grammatical and psychological interpretation as mutually dependent activities within the hermeneutic circle. Grammatical interpretation is the analysis of a text in terms of its structural and linguistic parts, with a view to establishing the accuracy and meaning of the text, through the use of methods that are essentially philological. Psychological interpretation, as Dilthey represented it, grasps the inner form of the text through “a projection into the inner creative process.”¹⁰ While the psychological and the grammatical are mutually necessary, Schleiermacher's famous distinction is generally seen as having given a certain priority to the psychological and thus as having created a hermeneutic that protects author and intention. While granting this priority, I shall argue that the “psychological” has been misrepresented and that Schleiermacher increasingly moves away from a concept of fixed intention. As importantly, this movement occurred because for Schleiermacher, unlike Wolf, textual criticism and hermeneutics were interactively rather than repressively related, thus creating a dialogue between the material and theoretical aspects of criticism that has been lost sight of until recently.

Schleiermacher's Critical Essay on the Gospel of St. Luke is a section-by-section “grammatical” reading of the text and of parallel sections in the other gospels that follows the disintegrationist methods of Wolf and Eichhorn. In his intermittent comparative study of the synoptic gospels, Schleiermacher questions the widely-held hypothesis of a lost “original” gospel, supported by Eichhorn and Alois Gratz, among others. In his microtextural account of the narrative and stylistic discontinuities in the Gospel of St. Luke itself, he steadily erodes the notion that the gospel was ever the creation of one man called Luke. Rather he sees it as a compilation of detached narratives from different sources, themselves sometimes compilations rather than eye-witness accounts. But since he is not committed to a hermeneutic that recovers authorial intention, he does not have to posit an original gospel that became corrupted or to construct an ideal text through the subtraction of problematic material, using what he criticises as the “comparative” method of eliminating what is not common to all the gospels and of selecting from them the lowest common denominator. The omissions and inconsistent repetitions in this layered composition are not always interpretively significant. Sometimes Schleiermacher will supply mechanical and historical explanations for them, related to the biographical circumstances of different eye-witnesses and compilers. As often he will see the completely different emphasis given an event in different accounts, both within the gospel and between the three synoptic gospels, as reflecting the fact that something is perceived in substantially different ways. In avoiding an overview of Luke and providing only a running commentary, Schleiermacher's criticism leaves itself open to theoretical constructions that it refuses to turn into dogma, allowing for the deconstruction of final intention as well as for the emergence of local intentions and ad hoc unities. Underlying his analysis, however, is a willingness to dispense with the critical fictions of ideal texts and transcendent authors. At the origin of the Bible is an oral tradition inhabited by relativity rather than governed by inspiration. The gospels are texts rather than books, collaborative attempts to piece together the life of Jesus. They are not intended as “regular books with a beginning and a conclusion,” and their “public authority” is due precisely to the fact that they are not “closed” to additions and revisions.¹¹

At first sight it will seem that Schleiermacher's anti-stemmatic textual criticism, denying as it does an originary text and thus precluding “a projection into the inner creative process,” is developed in connection with a text that has no Romantic or modern equivalent. For the synoptic gospels have a polygenous textual history. They are not written by a single author but transcribed and compiled by several people over an extended period of time and therefore cannot have a final intention. Although Plato's dialogues—also the object of Schleiermacher's textual commentary—seem different in that he treats Plato as their sole author, it is arguable that their particular assumptions about the dialogical and intersubjective nature of texts are also generated by a similar though less radical polygeny. Plato is in some sense the editor and compiler of Socrates' reflections, the original author being no more available to us than the original gospel postulated by Eichhorn. This raises the question of whether the deconstruction of final authorial or textual intention is unique to manuscript cultures as opposed to print cultures in which we generally have identifiable authors. Or to put it differently, are the models generated by studying manuscript cultures, and the “pre-autonomous” works that H. R. Jauss finds characteristic of them,¹² relevant to a criticism of texts produced in terms of a Romantic ideology of the self? I would suggest, however, that for Schleiermacher the field is not structured by an opposition between autonomous and pre-autonomous texts, any more than it is governed by a division between textual criticism and hermeneutics. His study of the gospels significantly influences his theoretical work in the period of the “Academy Addresses” and the final lectures on hermeneutics (1826-33). Inasmuch as the Hermeneutics assumes a single author for each text, we are therefore justified in extrapolating from a theory of how to read single authors—influenced by a textual study of composite authors—the rudiments of a revisionary textual criticism that Schleiermacher might have applied to the study of the nineteenth-century secular scriptures, had it been the convention to edit the works of one's contemporaries. Such a textual criticism, moreover, will radically call into question the ideology that underlies the theory of final intentions.

Schleiermacher's lectures on hermeneutics were developed over a period of some thirty years. In his first forays into the subject, he does not use the term “psychological” at all, preferring the term “technical interpretation” for a reading that is formalist rather than philological. Because in the 1819 Compendium he seems to substitute the term “psychological” for the term “technical,” he is often credited with mystifying authorial subjectivity, and it is assumed that this Romantic disease can only grow worse with age. What has not been recognized, however, is that in the final 1832 lectures he ceases to use the two terms interchangeably and actually distinguishes them. The term “psychological” becomes the site for a questioning of the unitary textual intention associated with technical craftsmanship and complicates the concept of the text as a verbal icon without replacing it with the concept of the author as sovereign subject.

This shift, of course, does not happen immediately, and when it does happen, it is more suggestive than categorical. In the first stage of this development, the introduction to the 1819 Compendium, Schleiermacher replaces the term “technical” with the term “psychological,” but then continues to use the former in a more detailed account that obviously does not work out the consequences of the new term. In the 1832 lectures on hermeneutics, however, Schleiermacher subdivides psychological interpretation into “technical” interpretation and “psychological” interpretation proper. Technical interpretation is concerned with the finished thought complex, where genuine psychological interpretation is concerned with the more inchoate thought-process itself. In recognizing that psychological interpretation involves the recovery not only of “leading” but also of “secondary” representations, of subtextual and collateral thoughts, Schleiermacher redefines intention so as to make it originally multiple and self-complicating.¹³ In other words, he transfers to the individual psyche the polygeny associated with composite texts. The psyche—which for Schleiermacher is linguistic and not ineffable—is already composite. This is not quite the same thing as saying that it is constructed out of various social texts, in the radically non-individual manner assumed by McGann as normative.

But if Schleiermacher remains “Romantic” in that his focus is individual rather than purely social, the Romantic ideology is already for him an ideology of the intersubjective rather than the unitary self. Crucial to his hermeneutics is the assumption that psychological reading recovers what he calls the “Reden” or discourse of a text. Influenced by Plato's dialogues, the concept of discourse posits a speech act that is not private property. Discourse is “the mediation of shareable thought,” a site for the interchange rather than the fixing of meaning.¹⁴ Characteristic of discourse, according to Paul Ricoeur, is the fact that it is addressed to another person and is constituted by an exchange of questions and answers, whether implicit or explicit.¹⁵ It is thus oriented to other possibilities that enter it as subtextual and collateral thoughts. The text considered as discourse, though not an avowedly collaborative enterprise as described by McGann in his discussion of Byron's relation with his publisher, is dialogic in Bakhtin's sense and is situated in a space that is cultural and historical as well as private. Nor is it possible to conceive of a text that is prediscursive, for even thought is essentially intersubjective and “becomes complete only through interior discourse.”¹⁶

This is not, of course, a theory of textual criticism, and textual criticism at the time did not deal with modern as opposed to classical works. But what we can say is that Schleiermacher's work on classical and biblical texts resulted in the emergence of a new direction in his hermeneutic theory, which, in turn, could have had significant consequences for the textual criticism of so-called autonomous literature. And given the integral relationship between hermeneutics and textual criticism in the late eighteenth and nineteenth centuries, it seems legitimate to draw out those consequences.

The editing of Romantic authors in the past three decades has made available a large amount of scholarly material in the form of notebooks, marginalia, and textual variants. One result of Schleiermacher's theory is a de-autonomizing of the individual text as it is studied in relation to such materials that enter it as subtextual thoughts. Textual criticism, which is not synonymous with editing, comprises the analysis as well as the collection of such materials, and there is of course nothing new about either activity, the use of historical and philological materials having always been a part of hermeneutics. What would be new would be the use of such materials to intertextualize rather than to contextualize the individual work, to disseminate rather than locate its meaning. Traditionally, the collection of documents has been a supplement to the primary editorial activity of establishing the outer form of a literary text and has helped to define its inner form by recreating its context. In a new textual criticism the editorial activity of recognizing that there are only versions of the text would be complemented by the collection and study of materials that disclose the diacritical nature of the text as part of a network of other writings. These materials reveal differences within the text by sensitizing us to “secondary representations” that a formalist reading might not track down.

A second consequence of Schleiermacher's work is a theory of revision that does not hierarchize the versions of a given text in relation to some ideal text, whether original, final, or eclectically constructed. Given the polysemantic nature of the text at the psychological level, the shaping of what Schleiermacher calls the “flowing, indefinite thought-process” into the “finished thought-complex” that occurs at the technical level¹⁷ through such finalizing acts as publication, can be no more than what Iser calls an act of illusion-making or schematization. Wordsworth's habit of deferring publication and replacing it with Dorothy's copying out of provisional manuscripts—a phenomenon we note particularly in the textual history of The Ruined Cottage—is one way of expressing unease with such schematization. Re-publication, the issuing of a text in different journals and arrangements of the author's work, similarly suggests the provisionality of publication. Obviously pre-texts—for instance the rudimentary fragments that comprise the first attempt at The Ruined Cottage—are not equal versions of the text, and the act of copying does mark the crossing of some kind of threshhold. Nevertheless, it is not possible to give authority to the published version of The Ruined Cottage in Book I of The Excursion over Ms. D., in which the stories of Margaret and the Pedlar are divided, or Ms. B., in which they are combined but outside the rhetorical framework of The Excursion. Moreover, a textual criticism that studies these versions cannot simply treat them as autonomous parallel texts but must also see them as intertexts, present within each other. Revision cannot occur unless hints of the new version are already present in the earlier one; nor can a revised text excise all traces of an earlier one, since aesthetic or thematic details that an author might want to change are not so neatly localized within individual words, phrases, or segments. Variant texts are part of a diacritical network in which, to adapt Derrida, each “element … [is] constituted on the basis of the trace within it of other elements of the chain or system” and is thus part of an “interweaving,” a “text produced only in the transformation of another text.”¹⁸

Finally, a third consequence of Schleiermacher's work would be a textual criticism that takes account of dialogism as well as difference: a de-autonomizing of the so-called final text not only in relation to other versions of the text or other documents by the author, but in relation to those other textualized voices—authorial or cultural—that enter it as subtextual and collateral thoughts.

We have talked so far about methods of textual commentary that might be extrapolated from Schleiermacher's work. His discussion of Luke, however, has a further bearing on the relationship between text and criticism. For Schleiermacher does not so much edit or interpret the gospel as show how the text and its transmission have become intertwined, how the boundary between text and commentary, author and editor, has become blurred.

The crossing of this boundary has two consequences. On the one hand, textual criticism can no longer claim to establish the material form of the text and, by a kind of synecdoche, its inner form or meaning. On the other, the text itself can no longer claim a transhistorical identity apart from editorial mediations that constitute and reconstitute it. For the theory of final intentions, in letting text and commentary be two autonomous discourses, allows the latter to determine the identity of the text but then also demands that it efface itself before the text whose true form it has revealed. In the late eighteenth century, people like Wolf and Eichhorn were committed to this autonomy of text and commentary that makes the textual critic the servant of the ideal text. But their theoretical norms were increasingly undermined by the empirical findings in what were then the three major areas of philological activity: the Bible, the Homeric writings, and the ballad tradition. For in none of these areas is an original text to be found. Schleiermacher, on the other hand, saw the text as enmeshed in its transmission, and thus challenged what Jauss calls “the substantialist idea of a self-contained work,”¹⁹ opening the way for a form of reception aesthetics.

We have commented on how the pre-publication history of certain Romantic texts reveals the intersubjective and multiple nature of intention. The case of Wordsworth dramatizes the implications of Schleiermacher's theory for the originary creative moment. William Godwin's editing of Mary Wollstonecraft's fragmentary Maria or The Wrongs of Woman offers a relevant contemporary example of how editing historicizes reading and therefore of the theory's consequences for reception.

If Godwin had chosen to efface himself from the text, he might have printed it without editorial interpolations and ended it with its last coherently written chapter, in which Maria's petition for a divorce is unjustly refused. In this case the novel would have seemed part of a genre of texts, including Keats' Hyperion, that are broken off at a crisis or climax. Godwin would have been “improving” the text, as Mary Shelley later did when she made Shelley's The Triumph of Life into a significant fragment by ending it with a question that sums up its title rather than allowing it to end aimlessly. The rhetorical effect of abruptly terminating the novel with the judge's unreasonable decision is to make us reassess his judgment in the court of future public opinion. But given the titular pairing of The Wrongs of Woman with A Vindication of the Rights of Woman and given the gaps between Wollstonecraft's feminism and her heroine's, the completion of the novel could not simply involve vindicating the petitioner but would entail reading beyond the gaps and silences in Maria's still sentimental feminism. The fragment, in other words, would function as an intentional whole, motivating the reader to articulate what it was historically precluded from saying as well as doing. As I have suggested elsewhere,²⁰ this strategic reading of incompleteness remains one approach to the novel, one which Wollstonecraft might have liked because it aligned her fiction with her political treatise. But Godwin is scholarly rather than sentimental in not ending the novel at its crisis and in printing various contradictory, almost incoherent notes for its continuation. The effect of these appendices is to complicate any psychological reconstruction of what the author intended and to make us recognize that her intentions are themselves a text that we must read symptomatically for its lapses and contradictions. Because some of the continuations are clearly autobiographical, their effect is also to make us intertextualize the text and the life, the text and the social and intellectual backgrounds. Godwin's additions consist only of a brief preface, an “advertisement” placed between chapters fourteen and fifteen outlining the projected plan of the whole, and a conclusion explaining the state of the fragmentary continuations. While they do not direct our interpretations, they place a frame around the text and locate us outside it, creating a historical space between author and editor and a further space between Godwin and us. We are not allowed to eliminate this space, which therefore becomes constitutive in our experience of the text. Precisely because of the novel's unusual status as a fragment issued posthumously, the text makes us aware of two matters of paradigmatic significance: how the activity of editing produces a hermeneutic thickening of a seemingly transparent text and how it places the text in an interpretive chain, where it is no longer the private property of either author or editor.

This effect of hermeneutic distance and relativity is specifically created by the fact that Godwin did not erase the activity of editing from the text and thus mute the text's historical status. In this respect he differs both from contemporaries of his who tried to improve an earlier text and eliminate its rough edges in accord with their own tastes and from the far more scholarly modern editors of the Bowers school who try to establish a definitive text and who assume that texts, to appropriate Pierre Macherey's distinction, are “created” rather than “produced.”²¹ Godwin was not of course a professional editor, but he belonged to the second of two schools of editing current in the eighteenth and early nineteenth centuries: one saw it as the editor's role to amend and modernize so as to create a reading text; the other required complete scholarly accuracy. Thomas Percy's experience in looking for a literary patron for the Reliques of Ancient English Poetry provides an apt illustration of the differences between the two editorial traditions. It seems that Samuel Johnson encouraged him to reproduce more or less exactly certain poems of antiquarian value from the Folio MS while Shenstone urged him to modernize the ballads and introduce “clarifying” alterations.²² The improver is motivated by the twin desires of restoring the author's intention by correcting the disfigurations and incoherences introduced by time and of creating a text that is readable in contemporary terms. Contradictory as these desires may seem, both aim to reduce the hermeneutic distance between author and reader: to create a text that we can read comfortably, in conformity with aesthetic norms of totality shared by author and reader. Though they differ in scholarly standards, the editor as improver shares with the modern editor committed to establishing the author's final intention a desire to make the best text possible and somehow to finalize this text in accord with the assumption that literature enjoys a special status among materialised objects. The ideological if not methodological connection between the editor as improver and the modern textual critic is apparent in the following statement by Tanselle on the editing of unfinished texts:

When writers leave unfinished, or unprepared for publication, literary works of other genres—those which are normally circulated in published form—… the rejected readings, false starts, and uncanceled variants are of interest in showing the author's manner of working and stylistic development, just as they are when found in the surviving manuscripts of a published work; but they do not reflect the essential nature of the work itself, as they do in a letter or journal. An editor who completes the author's job by preparing such works for conventional publication (correcting errors, choosing among uncanceled variants, and the like) is not obscuring the final effect or meaning of the work but rather clarifying it. When a poem, left in manuscript, is posthumously published in the form of an exact transcript, it is being treated like a historical document; when it is published in a clear reading text, it is being treated like a work of literary art.²³

In arguing for a reading text that disguises its editorial archeology, Tanselle assumes that literature has a special status. In presenting The Wrongs of Woman as a historical document (which is in no way incompatible with its being a work of art) Godwin follows a very different path: he allows us to see the text as produced under certain social and psychological circumstances, thus opening it to historical analysis as well as critical synthesis. As importantly—and this too is in the spirit of the new historicism practised by Schleiermacher—Godwin, through his preface and intermittent commentary, draws attention to the text as mediated. This effect—of the editor being part of the text rather than effacing himself before it—is partly a visual one, deriving from the fact that Godwin's comments are not relegated to the footnotes but are introduced in parentheses at various breaks in the text so as to mark the text as edited and reconstructed. Schleiermacher achieved a similar effect by producing not an edition of Luke (something which would assume that the text can be established) but a textual commentary longer than the text itself. For where a certain kind of edition makes the work of editing largely invisible by relegating it to an apparatus which does not have to interfere with our reading of the text, textual commentary substitutes the apparatus for the text. It thus focuses our attention on the essentially recursive process of how texts are established, withholding from us the final product of this process and thus the experience of reading something that is hermeneutically immediate.

We have so far focused on certain consequences of Schleiermacher's theory for an understanding of how texts are created and received. One further area remains to be touched on, and that is the study of textual criticism as an object in itself. Luke is not simply a text the date and authority of whose components Schleiermacher is engaged in determining. It is a text that exhibits in its archeology the process of editing and transmission. Although Schleiermacher does not actually consider textual criticism as a genre, as part of a textual and intertextual network, his shift of interest from what is said by the gospel, its signified, to how the text has been compiled by layers of editing, introduces a new self-reflexivity into textual commentary by making the activity of editing rather than the establishing of a text its subject. This kind of self-consciousness becomes much more explicit later in the century when writers like Carlyle and Kierkegaard thematize editing by introducing such figures as the editor of Sartor Resartus or Johannes Climacus. To play with the figure of the fictional editor is in some sense to introduce the idea of the editor as figure. For the incorporation in the text of the editor who supposedly superintends and legitimizes the text from without reveals the figurative structure of the literal activity of editing. The fictional effacement of the boundary between text and editor has a bearing on actual editors, such as Godwin as editor of Wollstonecraft or Mary Shelley as editor of Shelley: their work too becomes part of a text, implicated in textuality.

Such a foregrounding of the editorial activity has two consequences. First, the text is marked as text rather than as the material form of the transcendental signified revealed by editing. Second, and correspondingly, much of what we have said about the text can also be applied to editing and textual commentary. Whether tacitly or openly, textual criticism too is intertextual rather than final. It is inhabited by previous editions, introductions, and commentaries that complicate its finality and autonomy. And as a mediating activity that transmits the text to future readers and places the text within a chain of cultural signifiers, it is itself part of this chain, subject to being studied in terms of its own textual and cultural strategies.

We need, in conclusion, to broach the question of how practical the reorientation of textual criticism explored here is.²⁴ We distinguished earlier between two branches of textual criticism: the editorial function and an interpretive criticism minutely engaged (as practical criticism generally is not) with the primary and supplementary materials gathered by the editor. “Intertextual” editions that de-autonomize the individual text obviously cannot be produced unless we abandon the format of the book for some McLuhanesque cinema in which texts are juxtaposed in a spatialized format. The present writer neither proposes nor sees as necessary a merger between editing and concrete poetry. The suggestions made here pertain largely to the second aspect of textual criticism, that of commentary: both the commentary that accompanies scholarly editions and the subsequent commentary they facilitate. Particularly in the case of editions of certain types of nonfictional prose, such as Coleridge's Notebooks and Marginalia, there is nothing in existing editorial formats that precludes an intertextual reading. Indeed the chronological arrangement of the notebooks and the alphabetical arrangement of the marginalia both suspend any attempt to unify Coleridge's fragments by subject matter and thus to construct a text rather than a series of intertexts out of them. Marginalia and notebooks are of course unusual examples.

The case is somewhat different when we are dealing with writings involving the author's eventual intention of producing an autonomous text (and I use the word “intention” in its phenomenological as well as its normal sense). For a long time it has been the practice to insulate such a text from its variants and history, accounts of which are distributed into prefaces and notes microscopically printed at the back of an edition, and then evaporated entirely by the popular college edition. However, this practice is clearly beginning to change, and James Butler's edition of The Ruined Cottage, like other volumes in the Cornell Wordsworth, provides a relevant example.²⁵ Admittedly, Butler reconstructs a reading text of Ms. D. of The Ruined Cottage that purges “The Pedlar,” and this extraction of a very sizeable text previously entwined with The Ruined Cottage and still included as an addendum to it as Ms. D. results in a text of Ms. D. of The Ruined Cottage that is somewhat illusory in its cleanness. But the apparatus (photographic reproductions and transcriptions of the manuscripts) is printed immediately after the reading texts of Mss. B. and D., and in terms of layout and readability is not subordinated to them. In other words, the edition provides us with enough material to “situate” the interpretive choice represented by Butler's reconstruction of a reading text of The Ruined Cottage Ms. D., and allows us to see the reading text as having a heuristic rather than a canonical status. It goes without saying that interpretive choices are inescapable and that we can do no more than recognize them for what they are.

In the case of Wordsworth's poem, the textual interpreter will also want to refer to Jonathan Wordsworth's The Music of Humanity. Wordsworth's book reconstructs autonomous texts of Ms. D. The Ruined Cottage and “The Pedlar,” and has the advantage of recognizing that the latter bulks almost as large as the former in Ms. D. But it is erroneous in suggesting that there exists a clean text of “The Pedlar” that has put behind it the traces of Margaret's story and become a preparation for The Prelude.²⁶ Finally, we may also want to reconstruct insofar as possible a reading text of “The Pedlar, 1802,” which Butler does not provide, because Dorothy Wordsworth's complete transcription of it does not survive. At present we must recover this particular state of the text by extracting from and rearranging the Pedlar material in Ms. D., where chronological stages of Jonathan Wordsworth's “The Pedlar” and “The Pedlar, 1802” are intertwined. To begin this process of reconstruction is to recognize that “The Pedlar, 1802” is a substantially different poem from “The Pedlar” as printed by Jonathan Wordsworth, closer in spirit to The Excursion than to The Prelude. In combining such a reconstruction with the texts and transcriptions provided by Butler and Jonathan Wordsworth, we begin to get some sense of the intertextual nexus that is The Ruined Cottage. For the poem is a crossing-point for the bleak naturalism of some of the early poems, the egotistical sublimity of The Prelude, and the Romantic pietism of The Excursion. As such a crossing-point, moreover, it makes us aware of similar interdiscursive elements in the early poems, The Prelude, and The Excursion themselves.

To return to our original question, then, certain necessary changes in the format of scholarly editions are already occurring. What this paper suggests is that there may be benefits in rethinking the nature of commentary and the role of reading texts, especially for texts with an unusually tangled history, like The Ruined Cottage, where the extant documentation permits us to reconstruct that history. Although Butler's edition allows us to see the interpretive choices he makes as relative, the resources of the edition are by no means easy to use. The average critic is likely to ignore them, to take refuge in the reading texts he provides, and to treat these as end-products abstractible from the process of arriving at them recorded in the transcriptions of the manuscripts. Reading texts are enormously useful, as my plea for a reading text of “The Pedlar, 1802” concedes. Clearly one cannot make global comments about their limitations, since all texts are not as complex as The Ruined Cottage. But in such cases (which are more frequent than the average critic is perhaps aware), we need to recognize that reading texts are not definitive. They are heuristic conveniences: provisional schematizations dialectically related to the material in the critical apparatus, as publication or copying is to the compositional process that precedes it (and sometimes follows it). Though the reading text is not a uniquely modern concept, its hegemony can be traced to a combination of market and critical factors: the need to exercize some scholarly control over cheap paperback editions, along with the New Critical desire for a well-wrought text that appears timeless, free of historical or archeological markings. Perhaps the time has come to introduce into paperback editions used in universities some of the information about textual history and variants that we are too apt to dismiss as esoteric.²⁷

Equally important is the role of commentary in scholarly editions. Commentary obviously plays a crucial mediating role between editor and critic, and a commentary that is interpretive as well as descriptive might encourage the reader to make more use of the apparatus. There are, of course, dangers in abandoning a scholarly neutrality which, at the same time, consigns textual materials to a highly specialized terrain. To avoid such dangers, interpretation should not try to provide a simple gloss on the manuscripts, but should instead try to suggest the different and intersecting ways in which the manuscript evidence can be construed. This is perhaps a way of saying that the boundary between textual criticism and hermeneutics needs to be less rigid: that commentary needs to encourage reflection on such subjects as the nature of creative intention and the assumptions being made when a commentary construes evidence in certain ways. One way to make commentary more interpretive and yet more dialogical is to include in an edition critical essays by more than one person. Another method is to provide, as Schleiermacher did, a running commentary that does not follow the narrativized form of an article and does not subordinate local interpretations of evidence to an overall argument.

Notes

1. (Chicago: University of Chicago Press, 1983), pp. 8, 42. It should be clear that my own essay is very much in the spirit of McGann's enterprise and that my disagreement is only with his use of the term “Romantic ideology,” particularly in the literary-critical twin to A Critique, that bears that title (Chicago: University of Chicago Press, 1983). McGann has always been sensitive to the fact that textual studies in the late eighteenth- and the early nineteenth-centuries are a resource to which we will now want to return (Critique, pp. 12-13). In the early 1980s, however, he located Romantic resistance to the Romantic ideology in the prose rather than the poetry (Byron being a notable exception) and did not posit an integral and dialectical relationship between literature and textual studies. But since critics change their views, my objections to the notion of a “Romantic ideology” now increasingly refer to the uncritical acceptance of that term by others (particularly the New Historicists).

   McGann himself, in a more genuinely historical spirit, has begun to qualify his earlier location of Romantic literature in an ideological vacuum. Of particular interest is his essay on “The Ancient Mariner: The Meaning of Meanings,” which sees Coleridge's Rime as a culturally layered text, self-consciously aware of its interpretive layers (The Beauty of Inflections: Literary Investigations in Historical Method and Theory [Oxford: Clarendon Press, 1985], pp. 135-72). In this essay, however, McGann still sees the text as conditioned by an ideology that syncretizes interpretive layers so as to create a multeity in unity, and he makes the reading of the poem something that happens only through the response of a contemporary reader who situates the text's romantic ideology (a view with which I would tend to agree, if The Rime is taken as a specific and not a paradigmatic case). More recently in “The Idea of an Indeterminate Text: Blake's Bible of Hell and Dr. Alexander Geddes” (Studies in Romanticism, 25 [1986], 303-24), McGann does allow for the kind of reciprocal relationship between textual theory and Romantic texts that I assume here.

2. Lee Patterson, “The Logic of Textual Criticism and the Way of Genius,” in Textual Criticism and Literary Interpretation, ed. Jerome J. McGann (Chicago: University of Chicago Press, 1985), pp. 78ff.

3. The Prelude is an obvious case of deferred publication; so is The Borderers, not published till 1842, and Salisbury Plain, which Wordsworth thought of publishing at various points from 1794 onwards but did not bring out till 1842 when it appeared as Guilt and Sorrow. Yet another case is The Ruined Cottage, a version of which appeared in 1814 as Book I of The Excursion but which Wordsworth had also thought of publishing as an autonomous poem in its double-barrelled form, as a separate poem devoted purely to the story of Margaret, and as a separate poem devoted to the Pedlar. The most obvious example of recontextualizing is the massive redistribution of Lyrical Ballads into the categories Wordsworth created in 1815. But there are other examples: “There Was A Boy” was written in 1798, intended for the 1799 Prelude but not included, published in the Lyrical Ballads, absorbed into Book V of The Prelude (1805, 1850), and also published as one of the “Poems of the Imagination”; “Vaudracour and Julia” was included in the 1805 Prelude, then removed and published on its own in 1820; “The Simplon Pass” appears as part of The Prelude, but also separately; “The Discharged Soldier,” was written separately and then absorbed into The Prelude.

4. The Letters of William and Dorothy Wordsworth: The Later Years: 1921-1850, ed. Ernest de Selincourt, (Oxford: Clarendon Press, 1939), 3: 1987.

5. Quoted by Jared Curtis, Wordsworth's Experiments With Tradition: The Lyric Poems of 1802, With Texts of The Poems Based on Early Manuscripts (Ithaca: Cornell University Press, 1971). See Curtis's discussion, pp. 6-13.

6. F. A. Wolf, Prolegomena to Homer, trans. with introduction and notes by Anthony Grafton, Glenn W. Most, and James Zetzel (Princeton: Princeton University Press, 1985), pp. 43, 47.

7. From about 1807 to 1869; see Fr. Klaeber, ed., Beowulf and the Fight at Finnsburg, 3rd ed. (Boston: D. C. Heath, 1922), pp. c-cvii. I am grateful to George Clark for drawing my attention to the nineteenth-century editorial history of Beowulf.

8. Coleridge, MS Notebook HM 17299, p. 51 (in the Huntington Library).

9. Hans-Georg Gadamer, Truth and Method, trans. Garrett Barden and John Cumming (New York: Seabury Press, 1975), pp. 164-68, 174; Paul Ricoeur, Hermeneutics and the Human Sciences, trans. John B. Thompson (Cambridge: Cambridge University Press, 1981), pp. 46-48, 50-52; Richard Palmer, Hermeneutics (Evanston: Northwestern University Press, 1969), pp. 88-90. Dilthey was, of course, Schleiermacher's biographer and popularizer, but he distanced himself from “psychologism” in Pattern and Meaning in History: Thoughts on History and Society, ed. and trans. H. P. Rickman (New York: Harper and Row, 1951), p. 69.

10. Wilhelm Dilthey, “The Rise of Hermeneutics,” trans. Fredric Jameson, New Literary History, 3 (1971-72), 231.

11. Schleiermacher, A Critical Essay on the Gospel of St. Luke, trans. Connop Thirlwall (London: John Taylor, 1825), p. 15.

12. Jauss, “Theses on the Transition From the Aesthetics of Literary Works to a Theory of Aesthetic Experience,” in Interpretation of Narrative, ed. Mario Valdes and Owen Miller (Toronto: University of Toronto Press, 1978), pp. 145-47.

13. Hermeneutics: The Handwritten Manuscripts, ed. Heinz Kimmerle, trans. James Duke and Jack Forstman (Missoula, Montana: Scholars' Press, 1977), p. 154.

14. “The Hermeneutics: The Outline of the 1819 Lectures,” trans. Jan Wocjik and Roland Haas, New Literary History, 10 (1978), 2.

15. Ricoeur, Hermeneutics, pp. 133-34.

16. Schleiermacher, “The Hermeneutics,” p. 22

17. Hermeneutik und Kritik mit besonderen Beziehung auf das Neue Testament, ed. Friedrich Lücke (1838), in Sämmtliche Werke, erste Abtheilung, Zweiter Band (Berlin: G. Reimer, 1864), pp. 155-56.

18. Positions, trans. Alan Bass (Chicago: University of Chicago Press, 1981), p. 26

19. “Theses,” p. 137.

20. In “Reading the Secrets of the Political Novel: Godwin and Wollstonecraft,” Studies in Romanticism, 27 (Summer, 1988), 228-39.

21. A Theory of Literary Production, trans. Geoffrey Wall (London: Routledge and Kegan Paul, 1978), p. 67.

22. For an account of Percy's editing of The Reliques see Leah Dennis, “Thomas Percy, Antiquarian vs. Man of Taste,” PMLA [Publications of the Modern Language Association of America] 57 (1942), 140-54, and Albert Friedman, The Ballad Revival: Studies in the Influence of Popular on Sophisticated Poetry (Chicago: University of Chicago Press, 1961), pp. 192ff.

23. “The Editorial Problem of Final Authorial Intentions,” Studies in Bibliography, 17 (1976), 205-206.

24. This paper was originally read at the 1986 meeting of the Modern Language Association, in a session called “Text and Intertext I: New Directions in Textual Criticism.” I am grateful for the incisive comments made by the session respondent W. Speed Hill and have attempted to deal with some of the issues he raised here. My suggestions have been stimulated in part by McGann's seminal essay, “Some Forms of Critical Discourse” (Critical Inquiry, 11 (1985), 399-417). Here McGann argues against the hegemony of narrative in critical writing and for the use of forms (such as the “array” and the “dialectic”) that discourage the hypostatization of a single interpretation that occurs when we narrate what we say and then give it the status of something that has “happened” (even if only in critical time rather than actual time). My own view is that the array (an example being an entry in a descriptive bibliography) is too specialized a form, and I have accordingly turned to dialectical rather than narrative commentary as a way of mediating between the edited text and the common (academic) reader.

25. James Butler, ed., The Ruined Cottage and The Pedlar (Ithaca: Cornell University Press, 1979).

26. Jonathan Wordsworth, The Music of Humanity: A Critical Study of Wordsworth's “Ruined Cottage” (London: Nelson, 1969). Wordsworth's commitment to an autonomous text of “The Pedlar” is evident in the fact that he again reprints it in a genealogical series: The Pedlar Tintern Abbey The Two-Part Prelude (Cambridge: Cambridge University Press, 1985).

27. The Norton editors of The Prelude: 1799, 1805, 1850 (Jonathan Wordsworth, M. H. Abrams, and Stephen Gill) do precisely this.