University of Virginia Library

Conceptualisations for Procedures of Authorship
by
Klaus Hurlebusch [*]

In interpreting the mind of another, the scholar also always interprets his own; and the history of ideas is shaped by ideas as much as by their interpretation. Do these familiar insights, essential for a theoretic foundation of the humanities, also apply to textual scholars? In their endeavour for accuracy in editions, they are said to be more concerned with the letter than the spirit. Normally their "higher criticism" is only conveyed indirectly through editorial technique and in formulaic conclusions. This proper mode of editorial reticence to explicate an understanding of an author and an oeuvre, of literature and literary genre, may easily give an impression that, witnessing to self-imposed pedantry, the editor's work is merely mechanical. But appearances are deceptive. The study of opposing positions in the history of classical and modern textual scholarship teaches a different lesson. The most distinctive standpoints within textual criticism express—amongst other things—the philologist's varying conception of the author, and of his own role as interpreter. This is not surprising. In his task to establish a version of the text which is as authentic as possible, the textual critic often finds himself faced with complex situations holding a multi-valent potential for decision. This gives rise to the question whether and to what extent the individual scholar's faculty of judgement—that is: his perception of the author applied to concrete source situations—may or should be involved in the acts of decision. Positions are divided on this matter between those who fundamentally assent to conjecture and emendation, and thereby indirectly to a realisation of

their role as editors in terms of their image of the author; and those who, basically sceptical about textual interference, also consciously refrain from forming a particular image of the author, valuing instead an openness of approach to the author as to their own work. The question which of these attitudes is the correct one can never be answered conclusively. But exploring their grounds can secure a vital interest for the history of textual criticism, provided that this is seen as itself part of a history of ideas and not merely as a record of sample cases to guide editorial decisions. In the field of classical philology, Rudolf Pfeiffer has made a contribution to this manner of historical approach in his History of Classical Scholarship.[1] In the modern languages, where textual scholarship must generally deal with more complex transmissional situations, often incorporating, too, a genetic dimension, there have as yet been only isolated attempts of a similar kind.[2]

Along with textual criticism, there are two other areas of responsibility within editorial philology in which it is impossible to work profitably without a conception of the implications of the philological task in question. Firstly: the arrangement of texts and choice of versions; secondly: the evaluation and the presentation of an author's work process. Here, too, the editor confronts the question of his attitude to the author and to himself. The less he commits himself, and thus the more he keeps aloof from forming a consistent understanding of the author and of himself, the closer he may in fact come to the oeuvre and the work process as such—the author's as well as the editor's.

The following is an attempt to investigate the ways editors see their role, by analysing their conceptualisations of authorial writing. Instead of the usual pragmatic approach to editors' interpretations (are their text-genetical presentations in keeping with the evidence from extant sources? are they readable and illustrative?) we shall be concerned with the editor's work in the context of a history of ideas. The editorial enterprise is not seen instrumentally, as a means to an end, but as a form of expression meaningful in itself. The authorial work processes, in their turn, will be considered from the same angle, with a view to finding new keys to understanding them.

The critical edition—two contrasting concepts

In 1899 Wilhelm Dilthey gave a programmatic lecture under the heading "Archives for Literature." It was largely thanks to this lecture that manuscript bequests of poets and writers were approved as archive material and ultimately taken into state care.[3] Dilthey, Schleiermacher's biographer, regarded original manuscripts as an invaluable key to an understanding of their authors. He wanted to see them collected and researched

in literary archives. He felt a "human breath" emanating from "drafts, letters, notes"[4] and considered it the joint aim of the appreciative reader and the researching literary historian to comprehend them. "What would we give today to be able to read into the souls of Aeschylus or Plato, through such simple and direct statements" (p. 5). An author's manuscript remains could, as it were, fill in the blank spaces in an author's involuntary self-interpretation of his life philosophy, only partially rendered if "isolated and coolly presented in print" (p. 5). Yet however highly Dilthey assessed the cognitive value of literary manuscripts, for him they were not intrinsically significant—for instance, as witnesses to unknown versions of the text—but indirectly as a part of an entire body of evidence, from which the author's understanding of the world might be deduced. Manuscript archives, accordingly, were not viewed as a preparatory facility for editorial work; on the contrary, such collections would "considerably reduce" the need for printed publication (p. 16).

The central historical interest here is not in the work and its genesis, but in the author and his interpreter. It is ultimately the latter's concern—as reader and researcher—to comprehend the existing work of the author; to express this in terms of experience: "to acquaint himself better with his own self."[5] The author is thus possibly better understood by his interpreter than he understood himself, because he was not conscious of his own creative activity.[6] If it be the common goal of authors and interpreters to develop a durable and representative understanding of humanity on the basis of a shared experience of the fundamental conflict between an individual and a social identity, interpretive empathy, according to Dilthey, weighs decidedly on the social side of the scale. With regard to the affirming of one's own intrinsic existence, moreover, the positions of the artist and the interpreter are exchangeable. Dilthey hence always stressed the "homogeneity of intellectual life" ("Beiträge", p. 251). In his theory of the humanities, creative and interpretive activity correspond structurally, receptive contemplation taking precedence over action in both cases. This receptive contemplation is based in an unforced openness and receptivity for external and internal perceptions (e.g., memories)—a view formalised, for example, in Dilthey's much-quoted circular definition of experience—expression—comprehension. Since Dilthey sees the process by which literary expression is shaped as largely unconscious, he grants it little importance compared to actual experience. It is not the ability of expression, but merely the degree of that ability which distinguishes the author from the interpreter. Similarly, it is only the intensity of the powers of mind and intellect by which they differ. The decisive factor here is that literary production is viewed

from the perspective of the recipient, who wishes to open or entrust himself to the author. This attitude, profoundly influenced by Goethe and the spirit of his age, has had a sustained effect in German literary studies. It is also evident in the development of the modern Germanist edition—as illustrated by the Weimar Goethe edition and the Stuttgart Hölderlin edition (see below: The reception-oriented editorial concept).

A wholly opposite attitude to the process of literary production underlies the so-called "critique génétique" which has arisen in France since the 1970s and concerns itself with the textual genesis of literary works as evidenced in the original manuscripts.[7] It has developed out of French structuralist literary theory, with which it shares a methodical objectivism concentrated on exploring the text in its spatial/temporal dimensions. What distinguishes it from structuralism is a broadening of the field of observation and vision: behind the text as a closed and semeiological object, its perspective takes in the actual production of the text ("textualisation", "avant-texte", "le texte n'existe pas"); behind the written form ("l'écrit") it views the actual writing ("l'écriture"), behind the printed work the original manuscript.[8] This extension of the critical object to include the original handwritten documents is not only quantitative; it also affects the reader's or interpreter's attitude to the works, now enriched by a "third dimension",[9] to their authors, and of necessity thus ultimately to himself.

Exploring a work's textual genesis involves reconstructing the text chronologically and in a multitude of successive states and versions, each discrete and of only relative validity. The reader is thereby presented with a complexity of relations by which to read the text—he may select and combine them at will, exploring a textually structured field according to his combinatory and imaginative powers. "L'organisation du texte sur la feuille, les marginales, ajouts, renvois, less textes croisés [. . .] dedoublent les systemes de significations et multiplient par là les réseaux de lecture."[10] "Critique génétique" objectifies by philological means the tenet of reception aesthetics that in the act of reading the reader imaginatively shapes his own texts.[11] The intellectual appeal of this "approche génétique" to literary works is probably based upon the fact that, perhaps more than any other methodology in modern literary studies, it satisfies the need for subjective self-realisation in constructive activity, providing, as it does, a field for developing analytical techniques, serviceable terminologies or distinct methods of interpretation.[12] It is hence characterized by a tendency towards specialisation and critical self-sufficiency, a kind of "critique génétique pour la critique génétique". Indicative of this tendency is its abstract image of the author, removed into a sort of "instance écrivante".[13] "La notion même d'écrivain se dissout

dans l'incertitude"[14] —to such an extent that the concept appears applicable no longer just to single individuals, but also to collectives.[15] This view of the author as the subject of constructive writing operations limited only by the instrumental conditions of his productivity reflects the interpreter's understanding of his own self. It is a view not derived from an objectified image of man serving an inter-subjective exploration of the self as a community being, but from self-determined work in which the decisive affirmation of the individual is embodied symbolically. The extent and intensity of interest in the processes of production as such and their self-determining quality—for authors as well as for readers—is indicative of the strength of a will to self-realisation in the individual striving for liberation from pre-formed social models. From the point of view of intellectual history, it appears particularly significant that there is a common ground between modern critics and modern writers on this point. "On a déjà beaucoup parlé de cette conjonction entre une littérature qui traite de sa propre pratique et une critique qui explore les mécanismes du travail littéraire."[16] Both sides agree that a more suitable route to self-development can be found in productivity than in receptivity. The fact that "critique génétique" hitherto has been concerned mainly with the autographs of those authors who are regarded as the forerunners of modernism: Flaubert, Heine, Proust, Valéry, Zola,[17] also demonstrates this affinity.

In the German-speaking world there is, as yet, no line of scholarship comparable to "critique génétique." The Germanist literary scholars bent on systematically exploring the text genetics of literary manuscripts are predominantly editors. Their investigations are hence for the greater part guided by pragmatic interests. The perspectives that "critique génétique" develops independently of practical editing can be found within the German-speaking world in critical editions and in related works of editorial theory. Their main characteristics will be clarified with reference to the Goethe Academy edition and the Frankfurt Hölderlin edition (see below: The production-oriented editorial concept).

The decisive criterion by which to contrast the reception-oriented and the production-oriented editorial concepts lies in the difference of the image that in each the editor coins of the author. In the case of the reception-oriented concept, the author is understood as a person who, in his literary expression, establishes an inter-subjectively recognised identity. Here the author is mainly attributed with a characteristically deliberate, even teleological, attitude towards himself, and hence towards what he and others approve as constituent of a personal self. In respect of such identity, the author and the editor occupy reciprocally exchangeable positions: each is a literary producer and each is also the

discerning recipient of such production. Each sees himself as a self (1st person) from the perspective of another self (2nd person) and knows himself as the other (2nd person) for the other. In the production-oriented concept, by contrast, the image of the author is modelled to the assumption that his artistic intent is directed primarily towards his creative work and not towards his own self, which is in principle veiled in obscurity. In respect of this tendency towards an indeterminacy of the personal self, too, the positions of author and editor are reciprocally exchangeable; each relates to himself and to others as a subject (1st person) to the object (3rd person). The relationship of the editor and the author to each other is that of observer and observed creator of self-determined work.

I. The reception-oriented editorial concept

Regarding themselves as middlemen between the authors and the reading public, nineteenth-century critical editors exercised a remarkable reticence; they spoke little of their own activity. This is true of Karl Lachmann and Franz Muncker in the edition of Lessing's Complete Works; of Muncker in the "critical-historical" edition of Klopstock's Odes; and, above all, of the editors of the Weimar Goethe edition.

Precedence of text presentation over the critical apparatus

The Preface to this critical Goethe edition in 143 volumes extends to only seven pages, and the information about the numerous manuscripts and prints of the poems in the first volume occupies a mere six pages. The apparatus of variant readings is also maximally concise, being meant as no more than a "critical appendix".[18] The presentation of the texts as the author intended them finally to be given to the public was considered of greater importance than the description of manuscripts and prints or the details of variants and versions. The editors' reticence reflects their decision on what matter they considered more or less worth communicating to the reader. If one recognises two different editorial aims, each attainable on the basis of authentic and authorial records as preserved for modern authors: firstly, to reconstruct the shaping and reshaping process of literary works, and secondly, to define the particular form of a work with which the author wished in the final instance to be publicly identified, it is clear that the editors of the Weimar edition unambiguously gave precedence to the second. The processes of composing and revising a work were merely of historical antiquarian interest to them.[19] Generally cognisant of Goethe's attitude to his literary output as embodied in the last edition of his works which he himself supervised (the

so-called "Ausgabe letzter Hand"[20]), they considered it their moral duty to respect his final intentions almost unconditionally.

The serious lack of information within the apparatus, crucially insufficient details about the manuscripts and the selective and isolative rendering of the variant readings[21] have been explained as due to the fact that the "editors of the Lachmann school"[22] were still bound to the model of editorial technique in classical philology. There may be some truth in this, even though, if so, the method has not become outwardly effective—the apparatus, for instance, is not arranged in the form of footnotes. However, the explanation casts underserved aspersions on the pioneers of editing in modern German philology, making them appear as indiscriminate imitators. What brings them close to their colleagues in classical philology is a formal matter: the merely selective consideration of witnesses not chosen as editorial base texts. Yet the reasons for such selectivity differ. In the case of classical philology, the selection serves to justify the critically constituted text and is hence determined by its purpose. In modern editing, by contrast, the author's authentic or authorised text did not require any justification, and the selection of the variants was thus not linked to any objective aim, but was subjectively motivated. This explains the methodological aimlessness of the apparatus in the Weimar Goethe edition. It is determined by "the most respectful consideration for the wider circle of educated readers."[23] "Negligences or mere spelling errors" in the manuscripts are excluded as an unnecessary encumbrance; it is only variation which "is heard in the spoken language and which affects the syllable count and hence, in verse, the metre" that is considered worthy of record.[24] This guiding rule, it is true, is not observed consistently.[25] The apparatus of the Weimar edition has been called a "dead apparatus"[26] by one of its fiercest critics. With regard to individual volumes this is certainly an unfair criticism, but as an overall appraisal it does characterise the curiously functionless nature of the apparatus. The editors can hardly have remained unaware of the obvious difference between transmissional variants (the concern of classical philologists) and the genetic variants integral to the author's own manuscripts.[27] Yet they clearly were at a loss how to deal with these,[28] because they saw themselves primarily neither as textual critics nor as documentary historians, but as readers. This understanding of their role corresponds to their model of the author.

The model of the author in the 2nd person—the self-realising author

Editorial opinion in the early period of German philology held it to be the author's essential characteristic to be able to communicate his own self through his text, and not to draft and compose a text. According to

such understanding, the author is a subject oriented towards listeners and readers (including critics), and observing and assessing himself creatively from their perspective, i.e., in the 2nd person. The editor, in his role as reader and through his endeavour for supra-individual self-determination, represents as it were the other self, the author's partner, and vice versa. The reading stance common to both is evaluative, positively or negatively. The editor's understanding, then, of his own role is modelled on his assumptions about the author. If he views the author first and foremost as a judicious reader of his works, he does not in principle desire to be anything else. He considers everything that links author and reader to be more important than what separates them. If, on the other hand, he sees the author primarily as the producer of his texts, he conceives the editorial task ultimately only in terms of their reproduction from a common ground of subjective creativity. In either case, the mode in which the author is viewed is irreducibly voluntative and at best only approximately suited to given conditions of writing and transmission.

If the editor favours the model of the author in the 2nd person, it implicitly follows that he wishes the author to determine in which version and in which order his texts should be edited. In other words: he does not wish to give room to editorial judgement or discretion in these matters. The editor thus sees himself akin to an advocate, an executor of the author's will, before the forum of his readers. This is an attitude obviously determined by ethical principles, since it implies that the writer is taken absolutely seriously as a person and that his intention is respected as the highest editorial principle. The consequences of this kind of editorial ethos as they emerge in relation to the creative and revisional work of the author are: a limited interest in textual criticism and source description on the part of the editor, and a modal misinterpretation of the author's work.

The author's intention

The author's intention, as is well known, is a fundamental concept much debated in editorial theory.[29] By and large it seems that, despite the problems involved, it cannot be relinquished without negative consequences for critical editing. If it were renounced, if editors were to declare the very concept of authorial intention on principle to be outside their province, discernible expression of such intention would be entirely subject to editorial discretion, if not indeed arbitrariness.[30] Too little consideration however has usually been given to the fact that the notion of authorial intention can be applied in two different ways: as intentio recta (author in the 2nd person) and as intentio obliqua (author

in the 3rd person). It is the second way which is relevant to textual criticism. The question here is, whether a text or textual part or particle as transmitted is, or may be considered, intended (meant) by the author.

Editorially, the predicate "intended" may on principle be attributed to a specific textual finding or its emendation only when modified as "possibly intended".[31] Yet the "possibility" is subject to many degrees of certainty and probability or uncertainty and improbability, in much the same way as what a third party intends (or means) can only be known on the basis of everyday suppositions. The editor assumes responsibility for adjudicating intention in a manner different from that for his making descriptive statements which can be verified or falsified from the extant textual materials. Whereas the reasoning behind descriptive statements can be completely objectified, if need be with the help of explicit conceptual definitions and rules of procedure, this is not the case with assertions of textual criticism concerning the degree to which a transmitted text can be intended (meant). These are always only partly objectifiable, and the editor can only vouch for them on the basis of an understanding and knowledge which he has gained in his role as reader, interpreter and textual critic. Hence they always hold good only in relation to the editor's familiarity with the author's voluntative and linguistic behaviour.[32]

Whereas editorial theory has hitherto been dominated by the reciprocal editor/author relationship in the 1st and 2nd person—the reception-oriented editorial concept—or the one-directional editor/author relationship in the 1st and 3rd person—the production-oriented concept—the implications of the notion of authorial intention may be clarified from a perspective in which the author occupies the 2nd and 3rd person. The author's literary activity displays two sides: on the one hand it fulfills a specific purpose, the production of texts, and on the other hand it is a form of expression, a way of articulating the self in the text. The concept of authorial intention may be differentiated accordingly.

The author's intention to produce and communicate a text: what is meant and what is intended

Writing, as it transposes conceptualised linguistic utterances into the spatial-visual dimension, is basically an artificial application of the imagination. It necessarily has repercussions on the creative expression, since it renders transitory conceptions permanent. However, the representational mode of linguistic utterances is not altered in their inscription. Hence, writing is not to be equated with acts of direct communication in oral speech that cannot be revoked. This fundamental distinction is not infrequently overlooked,[33] which leads to a modal misinterpretation

of texts in autograph. Writing is without doubt an intentional act. Yet is the result of this act, the written text, intended by the author? The answer to this question is open where the author's attitude as reader to the text is indeterminable. Such is in principle the case with all texts he has not (or not yet) published. All that can be said about them is that the author intended them as possible results of his writing, or even that he potentially intended them as definitive and no longer disposable compositions. In respect to such texts, it is more profitable to rephrase the question and ask whether they, as they appear in written form, were meant, or could have been meant, by the author. Whenever one is involved in transposing conceptions into words or into writing, that is, when his own thoughts, and not those of his listeners or readers, are the central focus, it is relevant to ask what is actually meant by what is spoken or written or whether what is spoken or written actually represents what is meant (particularly in the case of an unusually or wrongly spelt word or one that does not seem to fit the context). What is meant is primarily that which is objectively intended. It represents the author's transitory textual resolution which could be explained further as that which he possibly wished to convey or present for reading by what he said or wrote.

The textual critic's scope for decision ends at the point where he cannot question what is meant by the author—for instance in the case of two separate autograph versions of a text. Here it is not within his arbitration, in his role as textual critic, which of the two should be reproduced. If the author has actually communicated one version, that is, released the contents and textual composition from his control and submitted it to another's (e.g., a publisher's, who has it printed and distributed), it is true to say that this version is intended by the author. But as what? Not as a text produced, but as a text received, also by others. The communication of a specific text version implies that the author, in his role as a reader, has positively assessed it, with a regard, too, to other readers. This means he has read and approved the text in the interest of other readers,[34] and in doing so he has asked himself whether his inner perception of man and the world (the yardstick of his identity and linguistic mode of utterance) is adequately reflected in it. Is the editor obliged to respect the author's intention to objectify his ideal self? Unlike the author's textual intention (i.e., what is meant), this represents, as it were, a higher degree of intention and is often singled out as the "real", "ultimate" or "final" intention. If this intention can be positively identified, then the editor is indeed obliged to respect it. It concerns, after all, that striving for self-determination and, eminently, for personal identity, to the expression of which literature above all lends itself. However, the editor's respect for the author's identity as objectified in

the literary mode does not relieve him of his duty to explore the genesis of the text and exercise his judgment as textual critic on the documents of transmission. It does not allow him to assume the attitude of a merely corrective reader.

The editors of the Weimar Goethe edition were of a different opinion. Choosing the "Ausgabe letzter Hand" as their basis, they believed the poet had taken the greatest care and had been conscientiously exact in producing a correct and final revised edition (WA I, 1, p. xix). Erroneously they presumed that Goethe's intention to have his works seen into print accurately with the help of an able philologist[35] had actually had a result corresponding to the intention. They did not recognise that the versions of the text printed in that edition, precisely because they had been painstakingly prepared verbatim et literatim, should editorially have been minutely re-examined to see whether, on that microphilological level, they were as intended by the poet. For it is true that, being the result of a work process, every state of a text, in manuscript or in print, only represents the intended state as a whole, and not in every detail. A text in toto (which means also: a text read) such as that of Goethe's final edition of his works is undoubtedly to be considered authorised by the author; yet this does not apply in partibus (meaning also: to the text as produced)—and especially not to passages where what the author meant can so easily be shown to have been subject to unintentional changes by himself or by his helpers.[36] The crucial question of textual criticism, namely whether a printed text is also the intended one, cannot be ignored a priori or narrowly interpreted with regard either to the accidentals of spelling and punctuation or to individual words. It is independent of the degree of authorisation in the preparatory phases of reproduction.

The editors of the Weimar edition exercised textual criticism with extreme reticence. They believed that they were basically entitled only to correct the final edition of Goethe's works, to eliminate obvious errors, but not to submit it to recension.[37] Valuing above all the edition's authenticity, they essentially did not regard it as the result of a work process, but as "the unmediated intimation of the author himself".[38] The editors' decision in favour of the final edition as the basis for their rendering of the text has its fundamental reason neither in textual criticism nor in historical research. The decisive factor for the Weimar editors was the author's self-determining will at the peak of his maturity, i.e., at the stage when his self-reflection in the mirror of his entire literary oeuvre was most consciously developed and his self-distancing ability with a regard for its readers was at its highest. Goethe's final edition represents the "pinnacle" of all his editions in his lifetime[39] in so

far as it attests his attitude towards the greater part of his literary oeuvre. It was obviously a complete matter of course for the Weimar editors to choose the poet's final edition and no other as the basis of their own.[40] This indicates that they were always more inclined towards a representation of the receptive-reflective Goethe than to Goethe as a literary producer. Their perception was of the poet who, conscious of his readers, affirmed or rejected what he had written, and not of the author who in writing would to a lesser degree have reflected the reader's perspective. Only from such partisanship could the poet's final edition have been accorded unquestioned canonicity. It is therefore no coincidence that evidence of the writer's work meant little to the editors of the Weimar Goethe edition, as shown in the critical apparatus. Likewise, it is no accident that they misinterpreted such evidence as the involuntary expression of Goethe's mind and intellect, as indicated in their qualifying the final edition of the works as "the unmediated intimation of the author himself".[41]

Disregard of the author's work due to modal misinterpretation

The great Stuttgart Hölderlin edition is held by German philologists to be exemplary of literary editions avoiding the information deficiency in the critical apparatus for which the Weimar Goethe edition is typical. The textual editor, Friedrich Beissner, records the manuscript variants in their entirety, including even the alterations irrelevant to the actual wording of the text; and he presents them readably. Without doubt, the critical apparatus is thereby significantly enhanced, even though it remains subordinated to the presentation of the text. As before, it is the writer's final and binding word that goes to constitute the edition text. The axiom of the reception-oriented editorial concept to honour "the poet's intention and his right to his work"[42] remains valid. In other words: the author is still granted the decision how his works are to be read. It is however debatable if this is possible in Hölderlin's case to the same extent as it appeared to be in Goethe's. For Hölderlin there exists neither a final revised edition nor "prints of individual works whose wording, since checked by the poet himself, should be binding for the editor".[43] Hölderlin's oeuvre has come to us largely as a legacy of unpublished manuscripts.[44] The editor's determination to let the author decide how his works should be read was therefore to be realised only by means of a text-genetical interpretation of the manuscripts, although these often enough are in the nature of heavily worked-over drafts.

Beissner's mode of text-genetical interpretation becomes fully apparent in his method and system of variant presentation. Its essential characteristic is to situate the variants within the context of a reconstruction

of the text's chronological development. The main focus is not on individual alterations, but on the representation of successive genetic versions. On the assumption that the author has replaced the earlier version by the later one, the later version is regarded as the one finally intended by him. If one accepts this system of text-genetic interpretation, Hölderlin's ultimate versions may indeed be constructed by stages from the heavily worked-over drafts. Yet is the implicit premise legitimate?

Beissner stressed repeatedly that his main concern was to develop from the "spatial confusion" of the manuscripts and their corrections a "chronological order of sequence" of genetic versions of the text. In this way he held that the genetic process could be rendered both presentable and comprehensible as a process.[45] The implicit consequences may be clarified by reflecting that the spatial dimension of an inscription, if irrelevant, can be considered so only from the reader's perspective but not from that of the writer. For the author, the spatiality of the writing importantly conditions his creative linguistic behaviour, enabling him constructively to negotiate his conceptions and their expression.[46] It is with the help of the spatial dimension that he renders his conceptualisations visible and available. He is in control of everything he has recorded in writing—which in turn may influence his imagination—as long as he is able to decide whether it should be retained or not. He can decide for as long as what he has written remains readable and decipherable to him. Deleted or otherwise invalidated passages must generally be taken to be included in such writing. Textual exclusions of this kind should therefore not be taken as definitive, i.e., as outright annulments, but as provisional deletions. Invalidated wordings within a manuscript must be considered potentially valid if still legible to the author. Conversely, all wordings left valid should be considered potentially invalid.

In his genetic representation of the manuscript texts Beissner sacrifices their spatial dimension. The alinear spatiality of draft initiations and alterations is turned into a linear temporality of text stages where each stage is definitively replaced by the next. This conversion is not a matter of mere externals. Rather, it reflects an assessment of the authenticity of the states of an unpublished text in autographs. The editor turns into definitive text what for the author in principle was only potential writing. Hence, the editor's system of genetic interpretation reinterprets the modality of the writing: from being author-related it becomes reader-related. Otherwise the editor's aim to make possible an "actively participating and co-productive view of the work in progress"[47] would indeed remain incomprehensible. It is in a manner the editor who produces genetic readings from the poet's draft fluidities. In so doing, he not only disregards the spatial dimension as something seemingly unimportant.

He also neglects the specific productive (constructive) function which writing has for the author as a process of visually recording transitory reflections that acts as a reinforcement of his imagination. He degrades, in effect, the act of writing to a mere external reproductive activity of no importance for its written result, which is paradoxical especially for an author whose manuscripts show that he made superbly constructive use of their spatial dimension in drafting his texts.[48]

Favouring the reader's perspective, this edition abandons the fundamental difference between the author's and the reader's evaluation of writing. For the reader, the act of writing as such is unimportant; important is only its outcome in what the writing imparts. Taking writing to be merely a graphic medium of communication, and categorically equating the written and the oral utterance, the reader views the act of writing as a graphic realisation of inner speech. This receptive stance is by extension attributed to the author when it is implicitly assumed that his characteristic and essential activity is to affirm or reject his writing in the reading of it as an articulation of inward perceptions whose content core remains unalterable.

In the Stuttgart edition, Hölderlin—like Goethe in the Weimar edition—is primarily understood as an author who positively or negatively evaluates his writing as an immediate projection of his imagined self. The manuscript deletions and changes are not considered in their merely potential quality holding no prejudice for the author's final decision on a given text in publication; they are considered definitive. A deleted or unaltered result of changes is not seen as the last of several textual options, but as the "only possible form" or the "consummate form" of expression.[49] The editor seems to have been aware that this reader-oriented system of interpretation is incommensurate with the nature of authentic textual production, since he justifies his text-genetical representation as setting out the "ideal growth" of the text and not its actual development, which "cannot generally be deduced from manuscripts".[50] This is an acceptable argument under the premise that it is relatively unimportant to know how the author worked, and more important that "organic links" should not be severed.[51] However, the "organic link" between states of textual development is finally seen to be the articulate author reaching out for the perfect expression of his self.[52]

The disregard and modal misinterpretation of the writer's work evident in the Weimar edition is thus by no means overcome in the Hölderlin edition, even though its textual editor understood much better how to deal with the manuscripts than did his colleagues of the Goethe edition (Beissner, "Editionsmethoden", p. 73).

To summarise: the reception-oriented editorial concept does not promote an interest in an improved understanding of literary work processes. Hence it develops no analytic system, differentiating terminology or representationally adequate model for dealing with text changes in autographs.[53] Philologists who favour this concept are more intent on preparing poet editions for the general reader than work editions for the scholarly user.[54] As far as possible, they eschew specialised editorial professionalism.[55]

As long as there are authors who sufficiently clearly lay down their decisions on the versions in which they wish their texts to be read, and as long as there are readers willing to submit their souls to authorial guidance, this editorial concept cannot be considered outdated.

II. The production-oriented editorial concept

Whereas Goethe's works were edited in the Weimar edition according to the reception-oriented concept, the first critical edition of Schiller's Complete Works of 1867-1876 provides an early example of the production-oriented editorial concept.[56] Some of its fundamental principles have been applied or at least adumbrated in this critical edition which is more concerned with a "history of Schiller's mind" (Schiller, 1, p. v) than with the Schiller who, within limits, determined the way in which his work should be read in the individual publications he instigated. (In a short life, it was not given to him to publish his collected works.) For his editors, Schiller's identity as he wanted it to be understood through his works was less important than what he produced in writing during his lifetime. Karl Goedeke, the editor, declared it his primary desire to document Schiller's creative mind in its individual artistic outpourings as set forth both diachronically and synchronically in their order of conception. It was his aim "faithfully to reproduce of each greater or smaller composition the earliest existing form, be it from a manuscript or printed source" (Schiller, 15, 1, p. v). Not the final authorised version but the earliest available one provided his base text. Moreover, since the aspect of authorial production took precedence over that of communicative self-determination, it was logical that the process of composition and revision should be reproduced as comprehensively and accurately as possible.[57] The record of Schiller's textual alterations (which, in fact, is not complete) shows the endeavour to present them not just as the results of alterations, i.e., deviations from the edited base text, but as acts of alteration with indications of how the changes were effected. This represents an important step in raising the status of the literary

work process, the writer's crafting shop as opposed to his self-sufficient 'thought workshop'.[58] How fundamental a step this was may be concluded from the editor's explicitly formulated insight that it carried consequences beyond the typographical conventions of text reproduction as designed for the readers. "Only photographic reproduction could give an idea of what the writer considered worthy of notation in the course of his work. Only photography, too, could clarify his manner of composition, which deleted letters and font variations do not suffice to illustrate. Yet it seemed an essential task to approximate the creative process as far as this may be visualised through the medium of print." This statement appears in the preface to the last volume of Goedeke's Schiller edition published in 1876 (15, 2, p. vi). Almost a century later, in 1975, the first volume of the Frankfurt Hölderlin edition drew the conclusion for editorial technique from such insight: facsimiles of Hölderlin's manuscripts complementing diplomatic transcriptions of their textual layout and a representation of the textual genesis became an integral part of the critical edition.[59]

For a long time the significance of the Schiller edition was not duly appreciated. It never became a model for methodology[60] and barely affected twentieth-century philologists who developed the production-oriented editorial concept.[61] Its most apparent innovation is a reconceived critical apparatus, yet this could not be developed from Goedeke's edition in which the apparatus was outwardly arranged according to the pattern of the apparatus criticus in classical philology. In modern editorial theory and practice the emphasis shifts from the reproduction of the text to the critical apparatus.

Equivalence or precedence of the apparatus over the text

If the editor sets the chronicler's approach above the reader's perspective, the author's published and unpublished work, the final and the earlier text versions, will rank equal with him. Goedeke was still undecided on this matter. Only Schiller's "youthful attempts" did he qualify wholly as "instances of his development and evidence of the history of Schiller's mind" (Schiller, 1, p. v). It was not in editorial Schiller scholarship but, remarkably, in a new critical edition of Goethe's works, the so-called Academy edition,[62] that this historical perspective came to be applied most decisively. This is particularly remarkable since for Goethe's works authorised publications exist in great numbers, a situation without parallel for Schiller or indeed any other author in modern German literature. It is a transmissional situation that does not exactly invite a levelling of ranks among the witnesses to the evolution of the oeuvre and its individual parts. The reasons for nevertheless adopting and consistently

upholding the historical perspective are mainly conceptual. The Academy edition, in truth, was designed to replace the section of the "literary works" in the Weimar edition.[63] It not only reflects an improved knowledge of the materials and a sharpened awareness of the problematics of textual criticism, it was also intended to convey a different perception of Goethe.[64] The "historical relief of Goethe's poetry, its outer metamorphosis which cannot be separated from its inner metamorphosis" was to be set against the "ahistorical nature" of the Weimar edition.[65] Ernst Grumach, the first editor-in-chief of the Academy edition, took his guidance no longer from the author's final communicated intention but rather from his original one: "It is only the author's original intention—as far as this can be deduced from the manuscripts and the first editions authorised by him—which can reliably inform the constitution of the text" ("Probleme", p. 45). In addition to its critical function of recording and justifying the constitution of the text, the critical apparatus was entrusted with the historical task of "relating a textual history" (ibid., p. 47) ranging from preparatory notes to the final version of the revised text.

Under Grumach's direction, the "historical relief" of Goethe's oeuvre was to retain its peaks and elevations in the shape of the critically constituted "best" texts in versions "adequately expressing the author's intention".[66] This constitution of the text was of course not to be achieved in the divinatory spirit of a Richard Bentley, but in the historical manner of a Karl Lachmann, taking into consideration "all extant textual witnesses and all factors which influenced the textual history" (ibid., p. 6). For this reason Grumach termed the Academy edition a "critical-historical" edition of Goethe's works.[67] The "historical relief" was levelled out by his successors, and the critical-historical edition became an "historical-critical" one.[68] They abandoned the aim of establishing a "best" text. The critical constitution of the text through recension lost considerable importance in favour of a greater respect for the preserved historical form of the text. Corruptions were to be determined only within the context of the base text and were, if possible, to be emended only in relation to a knowledge of that text. The process of recension was to be carried out on the basis of internal evidence only, without recourse to text-genetic knowledge about the author's habitual usages in the language and in his writing.[69] Consequently corruption was adjudicated within the limits of more or less obvious error.

This limitation in applying procedures of textual criticism resembles the considered restrictive practices of the Weimar Goethe edition: les extrêmes se touchent. A reserve towards textual criticism is common to editors who see themselves primarily as the author's loyal trustees and

those who see themselves preeminently as documentary historians of his work. Grumach's successors indeed only pursued the path he took to its logical end. In defining "the poet's original intention" as mainly binding for the reproduction of the text, Grumach implicitly adopted a critical stance that considered Goethe's creative behaviour as his essential characteristic. The authorised first edition, or indeed the manuscript behind it, seemed closest to the author's immediate production. "That which no longer appertains to the process of poetic composition and revision, the predominantly corrective, intended or unintended alterations in the later printings and editions" was relegated to the critical apparatus.[70]

With such partisanship for Goethe in his capacity as producer of texts, the differentiating potential inherent in terms such as "composition" and "revision" was apt to wane. To Grumach's successors, composition and revision, conceptual and revisional work, creative and corrective shaping were all indistinguishably aspects of a textual development resulting equivalently in text or text versions.[71] This in turn made the choice of the first authorised printing or its manuscript source as the base text for the edition questionable as such. It could only be maintained on extraneous grounds, such as an allegiance to the established practice of the Academy edition itself or to that of other editions.[72] At the same time, the basis dwindled for a more demanding type of textual criticism that would take the author's recognisable habits as a guideline, when possible, for the constitution of the text in the case of oversights or transmissional corruption. On the one hand, an attitude equating the author's published texts with the unpublished ones would imply seeing even the published ones essentially as the results of his creative work. Yet if, on the other hand, the published texts and versions are regarded as in essence not intended by the author, the textual critic is deprived of a reliable basis for his judgement. Should he, as editor, follow the linguistic and orthographic usage in the authorised prints, or should he adopt usages deduced from the manuscripts—even though these may vary to an extraordinary degree?

The restrictive attitude towards exercising textual criticism evident in the volumes of the Academy edition since 1963 is essentially the editors' own choice. This needs to be emphasised, since there exist also restrictions to textual criticism grounded in objective criteria, e.g., in the case of authors who largely did not publish their work,[73] or in the case of authors characteristically undecided about which of their texts should be published, and in which form; having perhaps written with no view to publication, extraneous and alien as they felt it to be to them. Such authors, to be encountered frequently since the Romantic Age,[74] present such a variety of possibilities of articulation that the editor often feels

acute incertitude about whom he is actually dealing with, what the author could have intended and what is to be accounted for as within his potential range. Yet a proper awareness of the author's perspective may still guide the editor towards an unbiassed perception of the textual conditions. Here, then, it is an author himself, as it were, who induces the restrictions of the editor's text-critical faculty of judgment. By accepting them, the editor respects the author's identity as manifest in his creative behaviour.

Given that every textual version, regardless of when it originated or how it was transmitted, may have an equal claim to being rendered in extenso, every version, too, deserves the same measure of editorial attention.[75] To the degree that authorised printings and authentic or authorised manuscripts are considered equal in respect of the text they document, their material differences gain significance. The material characteristics of the documents, both textual and bibliographical, which seemed only marginally relevant to the Weimar Goethe edition, have a central information value in the Academy edition. The detailed descriptions of the witnesses selected, which are contained in its apparatus volumes published since 1963, serve largely to justify both the record of texts and variants, and the genetic reconstruction. The record of the variants to the fully rendered text is not only complete, it is also presented by the same semeiographically exact method of transcription as the text itself. With regard to this uniform transcriptional method, the foreshortened record of variants in the apparatus ranks equal with the extensive presentation of the text.[76] The emphasis in both cases is on the attested text rather than the authorially intended one. In its main aim and method, the Academy edition is thus a document edition.[77] Moreover, with regard to the number of witnesses edited, each apparatus volume actually ranks higher than the corresponding text volume.[78]

Where a scholarly edition is presented principally as a document edition, the spatial dimension appears more important than the temporal dimension. The text is understood primarily as an autonomous and visually perceptible art object of a semeiographic nature, not as an author's aurally perceptible verbal utterance, as under the reception-oriented editorial concept. Siegfried Scheibe has expressed a sense of this modern editorial conception of the text with welcome clarity: "An editor's aptitude is revealed by his ability first to visualise a manuscript full of corrections as if blank and untouched, in order then to recognise and to infer, as far as possible, how, in discrete stages of composition, the blank pages were filled with the characters and symbols of a 'text', and how parts of this 'text' changed while others remained constant."[79] Scheibe also conceives of the genetic development of a literary work according

to such an inscriptional process of textual construction: the genetic stages in the work's development correspond to the textualised states of its pre-versions and versions (ibid., pp. 16-17, 28). This is a useful notion of text and text development in editorial terms, since it promotes a semeiographic analysis in close documentary harmony with the reproduction of the textual witnesses. In the Academy edition, it has proved profitable for the handling of the immediate witnesses to Goethe's work and that of his assistants, i.e., the manuscripts themselves. Their variants are rendered with the greatest attention to differentiating detail. Transcribing the manuscript variants with utmost exactitude, the edition may be said to approximate quite closely Goethe's working procedures; it certainly comes closer to them than does the Weimar edition in its selective apparatus of readings. To be more precise: in its mode of editorial presentation the Academy edition has been adapted to the author's working method by means of a differentiated rendering of the results of the work process: the acts of alteration are captured in their results, and the results in turn are viewed as integral to states or versions of the text. The author's work process is thus properly speaking only indirectly documented and does not itself become the subject of editorial presentation. Deleted passages and the structured spatial arrangement of the text (variants as spatially co-ordinated textual elements) are therefore not identified.[80] Scheibe's concept of text and text development serves in the Academy edition not only as an instrument of editorial analysis, but also as the aim of editorial presentation. The written traces of the author's working procedures seem worth accounting for only in as far as they lead to new versions of the text; what they may mean with regard to the author appears largely irrelevant. The dominant perspective is that of a reader who, provided with an edited text and variants, may combine, as in a puzzle, those versions of the text and strands of text development which at any given moment he desires. In gaining a sense of text production, he is thus brought closer to the author.

The model of the author in the 3rd person: the author concealing himself as a producer of texts

Introducing the radical historicism of some of the volumes in the Goethe Academy edition means assuming the perspective of the re-producing editor as normative. If all versions of the text which the author has produced or assisted in are declared equivalent in principle, it is wholly for the editor to decide which one he should reproduce in full. Authorial decisions in this matter have no force. The author becomes a subject without intersubjectively binding intentions if what distinguishes the texts he published from those he left unpublished is nothing but their

material state in prints and manuscripts. In other words: the obligation of respect for Goethe's decisions as to how he wished readers to understand his works and himself loses its validity. The editor's obligation towards the author is replaced by his obligation towards the scholarly users of critical editions.[81] The ethos of the—modern—editor is thus basically object-oriented, its concerns are the oeuvre and the text. Hence, the demands raised by the supporters of the production-oriented concept for greater objectivity in critical editing are largely, if not exclusively, to be understood as a desire to strengthen its text orientation.[82] They feel responsible mainly towards the users of editions to whom methodologically verifiable information is due about the author's texts and their transmission.[83] The author, on the whole, is reduced to the role of a producing supplier of texts and text versions which editorially creative skill re-produces for all manner of uses. Intentionality is envisaged mainly in the general sense that an author must be supposed to have intended to produce everything he wrote, which is tantamount to rendering the notion of authorial intention irrelevant.[84] The author's willed disposition of intention is conceded only in relation to isolated instances of more or less obvious error. These are also regarded, however, under a perspective that places the author in the position of a third person: is x, y, z intended by him, could it be expected of him?

It stands to the credit of the Academy edition that it has established a terminology providing the most discriminatory editorial instrument hitherto developed.[85] It bears the unmistakable imprint of a double focus on textual production (author) and textual re-production (editor/reader).[86] Two examples may serve to illustrate its inherent focal constraints. Firstly, the editors are ready to admit that they have not succeeded in articulating an "unequivocal" definition of "an author's work (=opus)" (ibid., pp. 16-17). The reason is that they have posed the question exclusively from a point of view of editorial text orientation and editorial concerns. Individual and authorially integrated texts of course present no real definition problems. Yet a definition becomes difficult if not ultimately impossible in cases where textual criteria do not exist, e.g., when an author has contributed to a collaborative publication or revised the texts of others for publication. Obviously his contributive pieces cannot be termed "works". Where should the line be drawn? Should self-contained additions be claimed for the "works"? Yet these may be related to revisions elsewhere for which the author is not himself responsible. The matter cannot be decided solely with reference to the author's producing role. His role as a reader-recipient needs also to be considered. Significant for the author is not only his "work", i.e., the text he has himself composed, but every text, whether written by himself

or not, which he can read and affirm as a representation of his life and his self and which he can thus adopt as his own. The corrective revisions for publication and ultimately the published versions themselves of texts by himself or others provide evidence of such authorial attitudes of identification. The author's willed identity appears ultimately disregarded if an edition takes indications of his judicious reading behaviour into full consideration only when recognisable from his self-produced texts but treats them selectively when found in the texts of others. Admittedly, to come into focus at all where only the productive side of the author is relevant, such indications of a judicious receptivity towards his own texts or those of others have to be text-genetically misinterpreted as evidence of productive activity, or else disregarded altogether. If the Weimar edition shows that its editors could make little of a Goethe self-absorbed in the production of his texts, yet much of a Goethe conscious in revision of how they should and would be read, the reverse is true of the editors of the Academy edition. By default of their methodology, they saw little to interest them in a Goethe revisionally shaping public texts.[87]

Secondly: the predominance of the idea of the author as a producer of texts has consequences for the concept of authorisation, which becomes directly or indirectly correlated to the author's productive behaviour. Accordingly, the term does not denote the author's affirmative reading of texts as expressions of permanence of his self. Rather, it classifies witness documents as authorially produced or co-produced (holographs or scribal copies with autograph corrections), or else as authorially induced or instigated (scribal copies, prints). Documents demonstrably originating directly or indirectly through authorial activity are regarded as authorised. The purpose of the concept is to distinguish the editorially relevant witnesses from the irrelevant ones.[88] The criterion, however, is helpful only in the case of the non-authentic documents, the scribal copies and prints; that autograph manuscripts are relevant witnesses is self-evident. The concept of 'authorisation' is in truth meaningless for variant selection, though not for genetic reconstruction of texts performed by editors and users of critical editions. The tendency to equate authorially written text and authorised text carries considerable consequences for the assessment of text versions derived from the author's manuscripts.[89]

Over-estimation of authorial composition due to modal misinterpretation

Within the terminology of the Goethe Academy edition the term "textual version" occupies a position of particular significance.[90] It is defined as follows: "Textual versions are completed or uncompleted executions

of a work which diverge. They are related to one another by textual identity and distinguished by their textual variance." Textual variance means a divergence in execution in terms of letters and punctuation. A revised manuscript may hence contain several textual versions (at the base level and the levels of revision). If such a manuscript is a holograph, all the versions contained within it are by definition authorised. What the author wrote is thus equated with what he intended. But what does this mean? Viewed logically, what is, and what can only be intended in this case is an adequate written expression of the author's imaginative perceptions. The distinctions between writing, expression and perception have disappeared, and writing appears equated with language representation. Whereas, under the reception-oriented editorial concept, writing is seen primarily from the reader's perspective as a merely reproductive act, it is the perspective of the writer activating his imaginative powers that dominate here. The author is viewed primarily as a producer recording his perceptions and, in doing so, evoking new ones. He is seen much less as a reader judging the written expression by whether it corresponds to his original perception or not. It follows that even that part of the author's work which he executed primarily in his capacity of reader is regarded as the production of new text. This new text, a version in relation to the earlier one, is by the editor judged to be a newly composed text which the author inscribed by means of variant notation merely to save labour. However, if textual versions really are newly composed texts that only accidentally appear in the form of variants, even these cannot really be variants, i.e., alternative possibilities of expression, but must be the definitive textual elements of distinct text versions.[91] Such modal misinterpretation of the variants as definitive components of authorial expression in new textual versions leads directly to the situation whereby the process of revision is understood as textual production and is consequently over-estimated.

It is a characteristic feature of the reception-oriented editorial concept to assume that what the author ultimately intends is an imagined and affirmable idea of himself as a person, in relation to which his creative ability of expression is secondary. Under the production-oriented editorial concept, the author appears in the inverse role: the primary regard is for the author's creative ability applied as boundlessly as possible; the will to commit ever-changing perceptions to inscriptional permanence is secondary. The notions of an accomplished expression and a perfected work can no longer be held; there remain only versions of the work in temporal succession that each represent new authorial attitudes. Accordingly, the identity of the author whose perceptions continually change finds expression not so much in the work as in the process of perception,

i.e., the author's working procedures. It is these that provide the real constant factor in the creative activity, and it was these that the editors aimed to bring to the fore in the Academy edition (ibid., p. 15), albeit in such a way that they might be understood as the means of textual production.

The Goethe Academy edition tellingly demonstrates that a preference for the author's productive rather than his receptive behaviour typically reflects a time-bound cultural self-image of those involved with literature. As will be clear, it is an attitude which is not suggested by Goethe's works themselves or their transmission, and which is far better suited to the works and manuscripts of more recent poets and writers who more often display a stronger leaning towards the productive function of writing than to the revisional one. In the case of such authors, versions of a work relating to each other in terms of identity and variance are usually scarcer than utterly divergent versions in the nature of independent texts.

Hölderlin's manuscripts also appear to reflect a primarily text-producing author. The Frankfurt Hölderlin edition is the only critical edition making a principle of an unrestricted presentation of the author's working procedures by means of manuscript facsimiles and diplomatic transcriptions ("typographic transcriptions"); at its core is a genetic reproduction ("linear representation") of the text. Faithful to its sources, and judiciously incorporating references to their graphics and topography where it synthesizes text by genetically conceived phase divisions, it nevertheless reflects Dieter Sattler's, its editor's, governing interest in the production and re-production of texts. Sattler himself draws attention to the fact that precisely this "process analysis" of manuscript findings and the consequent "textual constitution" are dependent upon editorial interpretation.[92] He also contributes to interpreting this interest in the author's productive behaviour in terms of a history of ideas when he declares for his own edition: "In its approach, it goes beyond the usual aims of literary scholarship in the same measure as it leaves a mental pigeon-holing behind [. . .]. Thus, if this poetry, achieved through self-sacrifice, is no longer given over to be enjoyed, if instead it serves an understanding of the necessity for individual and general opposition to 'imposed laws and their executors', the gain will actually depend less on the artificial end products than on the conditions and steps by which such thinking becomes aware of itself [. . .]. It would be worth relearning the power of thought which moves beyond its own self-sufficiency, the ultimate superiority of individual integrity over systems that survive only by calculated opportunism" (ibid., pp. 124-125).

This reveals a decided preference for a type of poet who develops

his ideas as independently as possible of considerations for others, of readers' expectations or their ability to understand his poetry, and who therefore has little need for self-distancing or self-censorship. It is an authorial condition preeminently attainable in the state of first writing down a text, yet less so in revising it. The genetic reconstruction of texts preserved in manuscript may hence be conceived as a means of penetrating towards the most freely imaginative author who, in this sense, is also the most individual one. His identity is often mysterious, indeterminable, to be reached out for only in such reconstruction. This model of the author reflects the self-image of the scholar definable by his interest in textual genesis. The quality which he values most in the author he claims also for himself, i.e., a minimal consideration for the community of readers and their wishes, which in turn means the greatest possible degree of individuation and boundlessness of self-being. Interest in supra-individual aims of authors is replaced by the greatest possible openness towards their reproducible procedures of composition.

III. The author's working procedures as a reflection of his sense of self-being

Editors who adhere to the production-oriented editorial concept consider it an important task of the critical edition to illustrate the author's working procedures, be it indirectly by documenting and describing the work's entire development and presenting the alterations it underwent, or be it directly by providing reproductions of working manuscripts. However, they do not acknowledge this process of illustrating working procedures as an end in itself. Its principal aim is to serve as a control of the editor's presentation of the text and its variants. This is also the primary function of the manuscript facsimiles and their "typographic transcriptions" for instance in the Frankfurt Hölderlin edition. Even their high degree of documentary authenticity in relation to the author's working procedures is really only to be understood as evidence corresponding to the meticulous reconstruction of the genetic textual conditions. Yet do the writer's working procedures merit observation and documentation only with regard to their text-genetic results? Are they of interest merely by their workshop handling of texts and variants, and not also as the author's medium of expression? In other words: are these procedures valuable only as a guide to new genetic states of greatest possible textual authenticity, and not in themselves as a mode of articulation of the writer who by their means created his texts? In one respect, that of revision, the answer must be in the affirmative. Writers, especially in earlier periods

of literary history, have repeatedly insisted how important critical reading and re-reading was for them and the consequent alterations of their texts. Pliny the Younger provides exemplary evidence in a letter justifying the written dissemination of his speeches to readers by describing with what critical care—and advice by friends—he revised and corrected the text of a speech: "Itaque nullum emendandi genus omitto. Ac primum quae scripsi mecum ipse pertracto; deinde duobus aut tribus lego; mox aliis trado adnotanda, notasque eorum, si dubito, cum uno rursus aut altero pensito; novissime pluribus recito, ac si quid mihi credis tunc acerrime emendo; nam tanto diligentius quanto sollicitius intendo. Optime autem reverentia pudor metus iudicant".[93] Acknowledging an habitual behaviour as important for himself, an author will also regard it as noteworthy in literary colleagues.[94] Lessing, for example, in reviewing Klopstock's Copenhagen edition of Messias, discusses its revisions at great length. As an author turned critic, he holds up many instances of Klopstock's textual changes for demonstration and gives a reason for the trouble he is taking: "Changes and improvements made by a poet like Klopstock in his works deserve not only to be noted, but to be studied with the greatest diligence. By studying them one can learn the finest rules of art; for what the art's masters well choose to observe attains to the force of rules."[95]

This was written in 1759, and thus at a time when literary good taste was still acknowledged by writers and readers alike as a bond securing a shared sense of the appropriate. The authors had the task of guiding taste in so far as they took it upon themselves to improve their own artistic sensibility by diligently studying master works of art, and by applying it in turn to their own works to improve the sensibilities of their readers. A writer who wished to be respected within the community of the literate did well to apply the highest degree of literary judgement to his works. The field to exercise his studied artistic sensibility was above all the redaction and revision of his own texts and not their conception or production. This exercise, however, was not a matter of "studium", but of "ingenium" or "genius", an outflow of the individual's powers of imagination and inspiration, and thus largely removed from conscious control. An author who ceaselessly improved his works in accordance with literary good taste thereby announced his claim to being counted among the better authors.[96] The extent of the revisions he undertook in relation to the actual conception and production therefore indicates the degree to which he saw himself as a community being and wished to be recognised as such by others.[97] Where this urge is about equal to the urge for creative individualisation, as for instance with Goethe or Schiller in their classic periods, the relationship also finds expression

in the working procedures. Given favourable conditions of transmission, it is relatively easy to distinguish procedural phases: the primarily productive phase of the preliminaries (notes, sketches, plans) and drafts (the drafting of individual segments, first integral inscription of a whole text) on the one hand, and the primarily revisional phase (fair copy, scribal copy, printer's copy, proofs) on the other hand.[98]

Such working procedures have a strong and purposeful forward direction even at an early developmental stage. The author begins relatively early after the initial inspiration to work towards an opus, even though he may be gaining clarity about its content and form only as it progresses. The purposefulness of his working procedures mirrors his teleological self-conception; his search for perfection in a literary work reflects his will to a perfection of his personal existence. Despite all possible individual distancing, his powers of imagination are bound to a traditional image of man which he accepts as the highest standard of his existence and self-being. It is not only his creative imagination which is more or less dominated by such self-determination; his entire work as an author is also so dominated. This means that his work is very much oriented towards publication. For the author can only relate positively to himself as a person if he succeeds in objectifying what he recognises as the essence of that self, raising it to permanence and providing for its unlimited internalisation by others. It is thus for the sake both of his literary and, above all, of his personal self-affirmation that this type of author views himself primarily as a community being. In this sense his consideration for the literate reading public is genuine.

To the extent that the regulatory power of an objectifiable image of man loses validity for the author's self-conception, his imaginative activity begins to develop its own, and often superior, dynamics. Correspondingly, the creative aspect (the recording, combination and evocation of perceptions) gains significance for his work, while the process of revision loses its functional essence. No longer limited to the revising of completed texts and text segments, it extends to the development of new texts and segments. Revision ceases to be clearly distinguishable from text production, which itself in turn, moreover, becomes increasingly characterised by an over-abundance of immediate, as opposed to retarded, changes. Recent German poets, e.g., Georg Heym, Georg Trakl, Franz Kafka,[99] offer numerous examples of such characteristics in their working materials.

Authorial working procedures directed primarily towards productive writing cannot provide a fruitful object for studying an author's artistic will in the sense of the Lessing quotation above.[100] They belong to a type of author less concerned with realising his artistic will or public recognition

as a superior writer than with individual self-realisation through the most boundless development of his imagination. The more he opens himself to his imagination and allows himself to be overwhelmed by it, the more his writing loses the quality of a purposeful finite activity divisible into phases such as preliminaries, drafts and revision. It develops a tendency, rather, of becoming a mode of existence with which the author exclusively identifies and in which he would wish to dwell for as long as possible. Baudelaire and Flaubert, those founding fathers of the literature of modernism, really did not wish to be anything but poets. Consequently, their pronouncements on literary work in truth bear witness to their individual lives and are meaningful only to themselves or their intimate friends. Baudelaire, the champion of "gouvernement de l'imagination",[101] records in a collection of notes: "Si tu travaillais tous les jours, la vie te serait plus supportable. Travaille [six] jours sans relâche. [. . .] Je suppose que j'attache ma destinée à un travail non interrompu de plusieurs heures."[102] Flaubert, who called himself "l'homme-plume",[103] often refers to his working procedures in letters. They bore unmistakable traits of obsession, sustained as they were by a "génie de la gestation, une sorte de jubilation, qu'on peut, si l'on s'en prend à l'homme, qualifier de folle ou de perverse."[104] In the course of composing Emma Bovary's seduction he commented: "Que cela continue! car je suis fatigué de mes lenteurs. Mais je redoute le réveil, les désillusions des pages recopiées! N'importe, bien ou mal, c'est une délicieuse chose que d'écrire, que de ne plus être soi, mais de circuler dans toute la création dont on parle."[105]

In a similar spirit Kafka records his mood of depression in situations when the imagination failed him in his writing, or that of an intoxicatingly inspiring extasy in moments when it ran free.[106] He appears himself to have sensed a spiritual affinity with Flaubert. Commenting on a few manuscript pages for L'Education sentimentale, he wrote: "The crossed-out pages [. . .] do not represent nights of failing energy. On the contrary, these are the very pages in which he wholly immersed himself, in which he lost himself to every human eye. And even when writing these pages for the third time, he experienced [. . .] this infinite happiness."[107] In his diary, Kafka articulated how much he considered productive writing as the only form of existence commensurate with his self: "When I became aware within my whole organism that writing was the most fruitful direction of my being, everything thronged towards it and rendered void all other abilities, directed towards the joys of sex, eating, drinking, philosophical reflection and, above all, music. I emaciated in all these directions."[108]

Writers who find their life's fulfilment in activating their imagination

by writing are, as a rule, scarcely able to adopt a clear stance towards the results of their writing. Their relationship to their texts is basically ambivalent. It tends to be positive as long as the texts provide a continued or renewed stimulus for the imagination, yet negative when the animation fails or cannot be experienced any longer.[109] Flaubert copied or rewrote passages from his novels or made "recopiages" (copies of pages of books he had read) to put himself into a productive mood: "Certes Flaubert lit pour se documenter, mais aussi et surtout par débauche imaginative, et ce sont les mots qui le debauchent."[110] And Kafka, clearly, was virtually incapable of substantially improving his short stories in revision. His ability of critical (re-)reading was little developed. Hence, his publications do not express decisions freely taken by himself and specifically relating only to his texts. They are determined, rather, by external factors alien to the author and to the text.[111]

The typical uncertainty of the authors mentioned with respect to what their texts might mean to readers taking a receptive attitude towards them signifies their fundamental lack of consideration for a readership altogether.[112] Such lack is not so much involuntary in the individual as a mode of intentional behaviour. The authors—and they are not only modern ones, as the example of Montaigne shows[113] —view themselves primarily as individuals and wish to be seen and accepted as such by others. Their daily existence becomes unbearable to the extent that this self-image is not acknowledged. In the same measure, literary writing assumes for them the significance of an authentic form of existence. Bearing no relation to real persons, it excludes the everyday world from its field of reference, and it is also creative, since the imagination can repress thoughts of an everyday community existence and blot them out in moments of happiness.

Documents of a mode of literary writing which, in fulfilling an author's wish to be understood preeminently as an individual, constitutes an essential articulation of his assumed form of life, are in this sense also biographic documents. This is increasingly true: the more significant that productive writing was for the author and the less important its textual results, the greater the indeterminacy or secrecy in his work or being, and the more clearly the work process shows signs of spontaneous inceptions or abrupt conclusions (indiscriminate use of available writing space, such as old letters, café menus, envelopes, pages torn from copybooks).[114] Such manuscript leaves body forth an author's innermost life sphere rather than his workshop activities. They are not merely the sources for "textologists" interested in reconstructing the genetic versions of the text, but witnesses sui generis which crave attention for their own sake.

Notes