Version classiqueVersion mobile

Text Genetics in Literary Modernism and Other Essays

Hans Walter Gabler

6. Theorizing the Digital Scholarly Edition

Texte intégral

  • 1 There is a conceptual distinction, and thus a terminological one, to be made between ‘text’ and ‘w (...)

1Endeavouring to conceptualise the digital scholarly edition, we may do well to begin by asking what a scholarly edition is taken to be in terms of orthodox principles. In general outline, a scholarly edition is the presentation of a text — literary, historical, philosophical, juridical — or of a work (mainly, a work of literature) in its often enough several texts, through the agency of an editor in lieu of the author of the text, or work.1 We see the editor as ‘agency’, functionary and guardian of the lifeline link between work (or text) and author.

2In support of the professional editorial function, the scholarly edition assembles several auxiliary sections of material around the text it presents. Interestingly, these are nowadays fashionably called ‘paratexts’ of an edition — a terminological borrowing from theorisings of the public staging of narrative fictions in books. Transferred to editions, it suggests that the edition text is seen as the core edifice, as properly ‘the edition’, whereas an edition volume’s other sections serve but as thresholds to it. The more traditional nomenclature is hardly less hierarchical, or divisionary. Here, the sections sensed as auxiliary are called ‘Apparatus’, ‘Annotations’, and ‘Commentary’. What apparatuses take care of, is comprehensively speaking textual material — or, let me better say: material text. Provided for users of the edition as edition, they contain the editor’s ammunition (as it were) for establishing the edited text in that editorial ‘agency’ endeavour of strengthening the author-text lifeline. Annotation and commentary, by contrast, support a second ‘agency’ function falling to editors, especially in the older traditions of editing: namely, one of mediating the text, or work, and of a text’s or work’s meaning, to readers. If we may so distinguish an author-text-directed from a meaning-and-reader-directed function of the scholarly edition, it is generally also true to say that, in the course of the twentieth century, the significant reinforcements in method and procedure of the former have led to a decline of the latter. Concurrently, however, the need for editions to declare themselves has been increasingly felt. Consequently, the series of their auxiliary sections has been extended by the Textual Introduction, incorporating an Editorial Rationale. Behind Textual Introduction and Editorial Rationale lies a claim that editions are not merely accomplishments of compilation, but essentially intellectual endeavours.

3Here lies our entry to outlining a revised model for the scholarly edition. What I want to explore is just what it may mean to view an edition not as an aggregate of parts, but as a coherent structure. The base line of my understanding of the scholarly edition is that it is a web of discourses. These discourses are interrelated and of equal standing. They are constituted, as discourses, by the editor, or team of editors, who provide as well as guarantee the edition’s coherence and intellectual focus. With their name or names, too, the editor or editors publicly assume responsibility for the construct of the edition as a whole.

  • 2 Time and again, to be sure, authors have themselves acted as editors of their works, and oeuvres. (...)

4Not an overly spectacular definition, perhaps. Yet looked at closely, it may be seen to turn the traditional sense of editions on its head by making not author and text, but the editor pivotal to an edition. This follows logically, to be sure, from the circumstance that, for close to two hundred years by now, the scholarly edition has been a product of academic learning and critical scholarship. It has thus always been offered to the public, general or academic, by the editor, never by the author of a text or work2 — albeit that editorial self-awareness has not always chimed with this logic. Editors have largely upheld a hand-maidenly bashfulness — a sense of the editor’s self-effacing role in the service of texts and authors thoroughly in tune with the estimation they were (and are) held in by the rest of the world. This cultural fact has had significant repercussions on the nature of the service provided and the work delivered. The self-estimation and cultural estimation of the editor has contributed importantly to hierarchising editions in the way described. The effect has been the dichotomizing of what editors have achieved and presented into work of theirs and work not theirs. Although the sections of editions deemed auxiliary were conceded to be their work, the work not theirs were the texts themselves, for the purpose of which the editions were prepared and existed.

5It is not to be disputed that the text an edition presents is not of the editor’s original composition and writing. That it is not, lies indeed in the very nature of the edition as a scholarly genre. Nonetheless, the text in an edition is in a real sense the editor’s text of the text or work edited. This, too, lies in the nature of the genre. Authors, as authors, would normally not dream of going public with texts or works under the tutelage of scholarly editors. Such editorial tutelage would claim too extensive an autonomy for an author’s comfort; for the scholarly editor always critically enters in between author, transmission and reader and always modifies the given texts as transmitted according to editorially self-defined principles, rules of procedure and practices. The edition text in a scholarly edition is hence always distinctly other than any one contingently historical manifestation of the text or work it represents. Being distinctly other in being itself historically contingent in its turn, the edition text in a scholarly edition is always the editor’s text.

6This, I suggest, is the foundation from which to rebuild a model of that product of scholarship and critical learning called ‘the scholarly edition’. A scholarly edition’s edition text is uniquely its text. It is, as the text of that given edition, the construction and responsibility of its editor, or editors. From this main proposition follows, too, that the parts which the edition, as edition, further requires, are not merely accretive add-ons to its text. They are the discourses that essentially ramify the edition text. That is, they are functional, being both related to the edition text and interrelated among each other.

7Among the parts, take, first, the apparatus reporting rejected readings or adopted readings from the copy-text and/or further collated witness texts. The apparatus entries specify emendations or refusals to emend and thus function as argument for the establishment of the edition text. Since the emendation apparatus thus argues the edited text, it stands to reason that the edited text must argue back: that is, must hold its own against the apparatus pronouncements addressing it. This confirms, reciprocally, the standing of the edition text as one distinct strand of discourse, and of the emendation apparatus as its ‘other’ — a discourse complementary to that of the edition text. Secondly, too, we discern the functional interrelationship of both these discourses with the (often separately listed) Historical Collation. The forms it takes depend on the outcome (as it were) of the exchange of reasoning between edition text and emendation apparatus. To this tripartite interchange may, fourth, enter the voice of the Textual Note, expanding and underscoring in natural language the argument implied in the emendation apparatus and historical collation entries. With these discourses in interaction, what we increasingly require is a key to what they are arguing about, and so determining. This key we draw, fifth, from Textual Introduction and Editorial Rationale. These five discourses — text, emendation apparatus, historical collation, textual notes, and textual introduction with the editorial rationale — play out among each other the game of the editor’s establishment of the edition text within the parameters of reasoning drawn from all available text material. The reasoning may furthermore need to go beyond assessment and adjudication of the material text. Arguments may need to build on meaning and require annotation; and annotation, expressed in natural language, may carry over into veritable commentary. Annotation and commentary, consequently, furnish an edition’s sixth and seventh strand of discourse.

8Being thus systematically, and even systemically, related and interrelated, these discourses have an equal standing within the construct termed ‘the scholarly edition’. Among an edition’s discourses, it is the Textual Introduction and Editorial Rationale that articulates its intellectual achievement. The relational discoursing within the edition’s system as a whole would (as already indicated) not in fact properly function without it. The interaction between the discourses of editorial rationale, text, and apparatus, in particular, supports the edition’s text-, work-, and author-directed functions. This is the area of strength in scholarly editions as we know them. By contrast, their user-and reader-directed functions: that is, their concerns grounded in content and meaning, and issuing in annotation and commentary, are in our day widely neglected. This is a serious shortcoming. For, shorn of its content-and meaning-related dimensions, the scholarly edition is one-sidedly textual and is simply not an edition for readers. Conceptually, therefore, the scholarly edition urgently needs to be rethought as again a functional whole. Annotation and commentary need to be brought back into it. For this to come about, however, these discourses should be brought back no longer as add-ons to the edition but instead as essential strands in an edition’s set of interrelated discourses, coequal with textual introduction, text, and apparatus, and interlinked with these not in serially additive arrangements, but in functional interdependence.

  • 3 The ‘knowledge site’ as a term has been given currency by Peter L. Shillingsburg in From Gutenberg (...)

9But this cannot nowadays sensibly be done in a book. The state of the art attained in commanding the digital environment enables scholarly editions, instead, to be ‘born digital’: the digital medium is becoming, and has become, the scholarly edition’s original medium. Hence, my contention is this: we read texts in their native print medium, that is, in books; but we study texts and works in editions — in editions that live in the digital medium. This opens up new opportunities for all criticism, scholarship and learning founded in material transmissions. The migration of the scholarly edition into the digital medium that we are experiencing in our day means that we may encounter scholarly editions in an autonomous environment suited in innovative ways to the study of these transmissions. Editions may in that environment be set up as complex instruments for exploration (‘machines’ for research, they might have been called in the eighteenth century). In such ways, they can offer us — as critics, historians, philosophers, cultural historians and analysts of texts, works and oeuvres — the novel opportunity of interlinked textual and contextual study in the multi-connectable virtuality of the digital medium. The otherness of the digital, as opposed to the material, medium will enable digital scholarly editions to reach out beyond the confines of the customary scholarly editions in book form. With a notion of editions as instruments of exploration to guide them, textual scholars as both critics and editors are put in a position of actually constructing and achieving editions accordingly. Digital editions may be designed and made researchable as relational webs of discourse, energised through the dynamics of the digital medium into genuine knowledge sites.3


10What could all this mean for the future work of scholarly editing? What implications, for instance, has the defining of an edition text expressly as the editor’s text? Generally speaking, it is a definition that puts the text of the scholarly edition on a par with the range of manifestations of the text or work to be found outside itself — on a par for example with the work’s manifestations made public in the text of its first edition; or in revisions of it; or in mediated editions to which the given edition offers an alternative. Ultimately, while the editor remains in undiminished duty bound to record and correlate all extant manifestations of the text and work (insofar as critically considered relevant) the editorial commitment is to the given edition as edition, and to the consistency of the rules established for it.

11For instance: authority, say (where determinable) or intention (where inferable, or actually evidenced, as may exceptionally be the case) are categories that the editor, in constructing the edition text, will ascertain, register and record. Yet the autonomy postulated for the edition text as the editor’s text of a scholarly edition will show them in a new perspective. They will cease to be an edition’s a priori determinants. Instead, they will function as an edition’s potential regulatives, to be actualised or not according to the editorial rationale. Whether or not they are chosen, that is, to regulate the edition, will be subject to critical reflection and decision. In terms of editorial procedure, this means that it will be deemed as acceptable and sound to mark ‘authority’ or ‘intention’ out as criteria for establishing an edition text, as it will be legitimate not to tie the editing to them.

12Not that this amounts to a hitherto unheard-of understanding. The clearer editors are, and have been, about what they are doing, the closer they come, or have come to it. Yet they still have universally had hovering over them notions of an absolute obligation to intention (if they are Anglo-American editors); or to authority or authorisation (which yet, on closer reflection, are document qualities, not strictly speaking properties of texts — or else emendation, let alone conjecture, would play no part in the editorial game); or, ultimately, to the author. Not to subsume the discipline of scholarly editing a priori to such externally adduced obligations, but instead to grant autonomy to the edition text as the editor’s text, will, in one respect, place squarely on the edition the onus of declaring in all specificity its internal regulatives. In another respect, it will free editors to undertake their task of editing in full responsibility and undivided loyalty to their declarations — a stance categorically different from editing in subservience to author, authority, intention, textual history, notions of received texts (the textus receptus of old), or whatever imposition of procedure else happens to rule the day.

13From this will in turn follow at least one all-important consequence. The editor, and the edition of the editor’s responsibility, will no longer decree a text. Edition and editor will instead, to the best of their rationale and ability, propose a solution to editorial problems inherent in a text, or a work and its texts. In terms of digital scholarly editions, this change in stance and attitude should prove truly momentous. As decreed, editions as we know them are, and have always been, basically static. Towards them, users as well as readers were and are exclusively at the receiving end. This pattern has so far, on the whole, been adopted, too, for editions that have already ventured into the digital medium. Their texts still tend to come in one shape, and one shape only. Being full-text-searchable, as they commonly are, they are yet mainly string-searchable, not relationally searchable; and their apparatus and annotation appurtenances are cumulatively incorporated, much like in the pre-digital times of the print medium. Thinking in terms of print adheres even to the (sometimes lavish) supply of images in such digital editions. Tellingly termed ‘digital facsimiles’ as they are, they tend to be ‘plugged-in’, so to speak, as another cumulative layer, and mainly as illustrations of a would-be materiality still hankered after. By and large, therefore, such digital editions are basically spill-overs from the print medium; they provide (generally speaking) an increase in comfort, yet betray little ground-breaking reconception. Stasis, ineluctably a feature of the material medium, has not ceded to the dynamics inherent in the digital medium.

  • 4 See Alois Pichler, ‘Towards the New Bergen Electronic Edition’, in Wittgenstein After His Nachlass(...)

14A significant reason for the survival of editorial thinking and procedure from the age of material print may be the persistent focus on the production side, on the making of editions. The user interface of digital editions has as yet been too little attended to. This may ultimately be due to the strong autocratic strain traditionally ingrained in the editorial enterprise. That strain effectively bars imagining the edition’s user as the editor’s partner and peer and makes for a lack of incentive to provide for the user’s participation in, and interaction with the edition. What would follow from the definition of the edition text as the editor’s text, by contrast, is (for instance) that the edition’s user, in turn, be in a position to interact with the edition text, and to interact with it both as critic and, in response to the edition’s recorded potential of textual alternatives, as reassessing supplementary editor (for the purposes, let us say, of generating alternative editor’s-texts from specific angles of interest, and for select use: say, to map out research pursuits along textual paths, or to support pedagogy-driven lines of enquiry in the classroom). The Bergen Wittgenstein edition, for more than two decades already a pioneer among digital editions, presently envisages yet once more to re-emerge, and this time in a mode of ‘interactive dynamic editing’.4 The technical means exist and can be made available with relative ease, as we know, for attaching satellite work platforms to the main research platform of a dynamically conceived and constructed digital edition. The challenge is to conceive and construct the edition platform itself as the satellites’ docking platform.


15To relate such a vision to pragmatic realities may give initial cues for considering just what it would take, and what it would mean, to design apparatus, annotations, commentary and, beyond that, document visualisations (commonly: document facsimiles), too, into a digital edition; and, furthermore, to strive for doing so not cumulatively, but relationally.

16The monumental Danish edition of Søren Kierkegaards Skrifter, while edited initially in book form, has meanwhile been thoroughly converted into a net-based digital edition. From it, we may view a first, and as yet cumulative, example of digital presentation.5 What it provides is a click on/off facility for every individual textual variant and annotation (‘Tekstkritik’ and ‘Kommentarer’ in the tool bar). These are arranged serially along the progression of the text. To be sure, there are already a few gestures in this digital staging of the edition that point to a relational use: clicking to the commentary asterisk at any pre-identified instance of annotation-and-commentary occurrence, one is given the opportunity to proceed further — or recede deeper into the edition’s storage (database) recesses. Over and above an instance-by-instance progress, one may also read a textual introduction, as well as consult a guide to other addressable categories of information and follow these up. On the one hand, all this provides a grid of categories serving purposes not merely of reading, but of using the edition. In this respect, the digital Kierkegaard edition begins to be organised relationally. On the other hand, though, the categories given are formal categories throughout, arranged either numerically or alphabetically as registers or indexes. This is helpful, and the result is useful. The digital presentation, storage and linking makes for nimble consultation of the edition and fast information retrieval in terms of the editors’, that is: the editorial team’s, decisions on what information digitally to identify and pre-organise for retrieval. One can use this digital edition just like any well-made edition in book form. The lure is that one can use it faster and with more versatile combinations of the pre-organised information. Insidiously, nonetheless, one remains cast as user to work with an edition still in (simulated) book form.

  • 6 What has been realised for Søren Kierkegaards Skrifter was, at the time of writing the present ess (...)

17The digital Kierkegaard edition is a fine example of an edition originally devised and realised as a book edition, before being migrated into the digital environment.6 It is worth reflecting a moment longer on what its click-on/click-off facility of textual and commentary annotation provides — and from pin-pointing what it does not provide, to widen the horizon of the potentialities of born-digital editions. ‘Click-off’ allows us to read a clear text. ‘Click-on’ opens windows in which in turn we are invited to read, and to read the notes therein contained either individually or consecutively. Consecutively in the list of variants, we may at will make something for ourselves to see the individual entries collocated. Similarly, we are invited to read the commentary paragraphs in sequence. What we cannot do in either of the edition’s sub-sections ‘Tekstkritik’ and ‘Kommentarer’ is to explore them in relation to one another, or in correlation, comprehensively, to the edition text.

  • 7 Back in 2002, at the annual conference of the Association of Literary and Linguistic Computing hel (...)

18What would it take to make such relationality possible? It would, I submit, be a large undertaking, and something new for editions — yet also something that would exponentially increase their usefulness. It would amount, in its turn, to a labour of editing on their own the apparatus and commentary materials in an edition — that is, of bringing into a genuine strand of discourse within themselves the materials as compiled to annotate and to provide commentary for the text or work offered in the edition. In such an operation of second-order editing (‘second-order’, that is, within the construct of the given scholarly edition as a whole), the content of the annotations and commentary would be treated as articulated text in its own right. Submitting the apparatus and commentary contents to grid connectivities and search patternings would allow bi-directional linkings, say, of any given note with any other note of correlatable content within the body of the edition, as well as with the edition’s backbone, the edition text. As this would imply assessment and evaluation of the note material as generated from the edition text in the first place, it would mean grafting a process of editing in a critical mode, a mode of criticism, onto the basic editing in a textual mode—the inherited mode of editorial scholarship. It would establish within the edition as a whole an expressly critical perspective on the text and work edited.7


19Considerable energy is expended these days, as we know, on incorporating document visualisations into digital editions. In this, what needs to be eschewed in the digital edition is replicating the facsimile-in-book, static in nature, of manuscript page images. As images, even as digital images, it is true, facsimiles are reproductions of document pages. Yet the static inertia that this implies need not be retained across the media divide. It is, for one thing, certainly not conducive to integrating document visualisations relationally, and in this to do proper justice to the often spectacular graphic heterogeneity of documents.

  • 8 The standard workaround practice I happen to be aware of that has meanwhile established itself for (...)

20Operations to overcome the inertia of facsimiles go, as we know, in the direction of sub-segmenting the facsimile page unit—something that cannot be had in the realm of material documents. But since a digital facsimile can be virtually cut up into even the minutest subunits of inscription, these can be individually sensitised for links (both mono-directional and bi-directional) to renderings in transcription and to digital text, as well as to annotation and commentary presentations. Conversely, attention ought to be given in future, too, to what might be termed a super-segmenting of document facsimiles. What is written on, or in, documents laid out in folios as codices, commonly bridges page divisions — be it the individual paragraph that begins on one page and continues on the next page, or the chapter extending over multiple document pages and folios. Where we have seen document contents in terms of text and, in terms of editions on paper, lifted that text off its material substratum, we have been used to subordinating the record of page divisions, or indeed to suppressing it entirely: the original ‘text carrier’ has often (as said) vanished in the process. But, as we gain the ability to uphold the visual interest in original documents through digitally imaging them, their division into material (folio and page) units makes itself insistently felt anew, and causes problems. Hence, procedures will need to be devised, in the virtualised representation of documents, to ‘overwrite’ their real-world material divisions. A seminal idea here might be to model a perspective by which document and document contents could be defined as co-extensive (and so delimited as a chapter, say; or a sequence of paragraphs singled out for analysis; or even just a paragraph extending over a page divide). The codex folio and page, as bibliographic units only, do not provide such coextensivity. But to ‘virtualise’ any given or pre-chosen set of codex folios or pages and to delimit it, in terms of its content, as a continuous ‘scroll’, might provide a viable solution for supra-segmentation ‘overwriting’ material page divisions (which yet, as such, would not, of course, go unrecorded as digital information attributes).8

21Interlinked into an edition as a whole, not only may facsimiles and facsimile sub-or supra-segments participate dynamically in an edition’s multiple discoursing but more importantly, perhaps, we may no longer think of document visualisations as somehow mere illustrative add-ons to editions. They may instead be properly recognised as constituting a core element of our editorial objects themselves. For what they represent is the genuinely other manifestation, the visual face, of the textual (and, as it were, logocentric) renderings of the transmissions edited. So to revise our awareness of document visualisations has a bearing especially on the editorial sub-discipline (if I so may call it) of editing draft manuscripts. These are by nature spatial, and thus to be apprehended visually before textual sense is (and often even before textual sense can be) made of them.

22Paper and the book are, and will always remain, the originating and the palpable factor in editing texts in public transmission, whether from medieval manuscripts or from print. It is for this vast range of our cultural heritage that the vision holds (as expressed above), namely that, while we will continue to read books, we may look forward to studying editions that live in the digital medium. To this however must now be added my further contention that manuscript editing, that is, the editing of manuscripts from private transmission such as notes and notebooks, drafts, diaries or letters, belongs exclusively in the digital medium, since it can only there be exercised comprehensively. This is so because from out of the potential inherent in the virtuality of the medium it is alone possible to encompass and represent the multiply double nature of private-transmission manuscripts. First, what such manuscripts contain and present is double-natured: it is processual, since it is writing; and it is, or points to, the result of writing processes, namely text. In another respect, private-transmission manuscripts are double-natured as documents: in a medial dimension, they function as carriers of text requiring to be read; yet at the same time, they have a material existence and an autograph quality for which they must be seen. Third, what text they carry is evident, that is: can be read directly from them; yet the meaning of the traces which the writing processes have left on them, and the interplay of those traces with the document materiality itself, can be elicited, if at all, by inference only and hence will always require critical interpretation.

23The task of editing draft manuscripts, consequently, combines the editing of such writing processes with the editing of the sequential textings or texts resulting from them. Thus double-pronged, manuscript editing in the digital medium constitutes a fundamental extension of the modes of scholarly editing. For it is in the digital medium only that the imaging of the third dimension of the manuscript space by means of link-related document visualisations may be realised. This, in turn, renders possible, too, an illusioning of the fourth dimension of time which is implied in the successive filling-in of the manuscript space. Consequently, manuscript editing performed in what we now may recognise and embrace as its native medium allows the experiencing of the processes, and simultaneously of the results, of writing and of texting in manuscripts.

  • 9 It was the experience of digitally editing draft manuscripts in particular that led me to propose (...)

24If these are theoretical tenets, they have exciting critical consequences. Manuscript editing in the digital medium is a superior base of operation, for instance, for genetic criticism. Drafts supply not merely textual evidence. They also, at times, provide evidence for the given author’s processes of thought behind the writing. To elicit such evidence presupposes a distinct interpretation of the material drafts, and the writing and writing patterns on them. Invoking the notion of ‘interpretation’ should cause no anxieties. For, rightly regarded, our acts of editing always require an admixture of interpretation and interpretive criticism. To edit a draft simply involves extending that requirement. In drafts, we interpret not only groupings of letters as words and thus read them as texts. We also chart and analyse visually the spatial patterning of the writing, interpret its likely temporal sequence, and interpret what we make out as ‘text’ in relation to both patterning and sequence. Since we cannot thereupon express our findings in other than our own words, the analysis and editing of a draft manuscript translates by nature (as it were) into presentation and representation closely interlaced with critical commentary.9


  • 10 Their faultlines in theory notwithstanding, the glory of the twentieth-century bibliographical way(...)

25With the advent of the digital medium for scholarly editions to live in, as we have observed, the lifting of texts off their material ‘text carriers’ no longer leads inescapably to their renewed material reinscription. The editorial object is set free for study in the logical, as well as virtual, digital space. Digital editions must however, in their turn, and precisely in their ‘otherness’, derive bearings from their texts’ native transmissions in the material medium. This is where recognising the primacy of the document — meaning that texts are, logically, always functions of the documents transmitting them — becomes essential. It is exactly where, and when, the text is and remains separate from the material support of its transmission that the material parameters of that support need to be adjudicated as potential determinants for the digital edition. To see the text fundamentally as a function of the document helps to recognise afresh that in all transmission and all editing, texts are and, if properly recognised, always have been constructed from documents. To edit texts critically means, precisely, to construct them. Conversely, the constructed texts of editions are in essence the products of criticism. This is as true in the essentially two-dimensional medium of paper and the book as it is in the virtual, multi-dimensional digital medium. Therefore, in theorizing the digital edition of the future, we need to account, too, for the critical dimension, indeed the critical nature of the editorial enterprise and its outcome in the scholarly edition.10

26To recognise textual criticism as constituent of criticism implies accepting as well that textual criticism is not synonymous with scholarly editing. The text of a scholarly edition represents, as one might say, the distillate of textual criticism well exercised. Yet, as I have argued, it forms only one distinct strand in the composite of discourses that make up the complex total of the genre of scholarly writing we call ‘the scholarly edition’. Text and edition are not coextensive. In its composite complexity, rather, the scholarly edition as a whole is both the product and the facilitator of scholarship and criticism. It is an instrument to organise knowledge through aggregating and thickening the givens of transmissions and their historically variable textual as well as contextual shapes and understandings through reception. Being that instrument, it enables analysis and generates knowledge in continuity, too, from the multiple of discourses that in total it organises.

27Yet, if such definitions and claims give us a reference frame within which to explore the nature and role of the scholarly edition for the twenty-first century, how is it that, in general awareness, editions are still mainly perceived simply as texts, and in terms of the texts only that they offer? Looking at the matter historically may help us understand the contingencies of our present perceptions. One determinant to be singled out is that the one correlation I am insisting on, namely that editing and editions are (systemically) a function of textual criticism, and a partial one at that, appears historically to have originated in the reverse. In the oldest traditions, editing was the prime cultural activity, and it was only in its service that procedures of thought and logic were invented and so devised as to systematise the pragmatic exercise of editing. That the ramifications of logical thinking amounted to a reversal in the functional relationship of textual criticism and editing has not always been thought through in its full implications, let alone been of concern to the average editor exercising ‘textual criticism’ pragmatically. Only exceptionally, too, does it enter the awareness of the readers and users of editions: they — we — only seldom stop to reflect what text-critical penetration and judgement is assumed and may be required to assess and appreciate that stuff which editions at their surface offer as texts. Lurking behind such a narrowing of awareness lies, I presume, that division of ‘criticism’ into ‘lower’ and ‘higher’ that we owe to (German) philosophy of the early nineteenth century, and which, through various permutations, brought to the Humanities in the twentieth century, and most insistently so in the Anglo-American spheres of intellectual hegemony, that strange, yet influential, dichotomy of ‘scholarship’ versus ‘criticism’.

28Ultimately, though, at the core of all such stances and constructions lies the need to stay mutability and countermand the loss to cultural memory inherent in transmissions — that existential human need, in other words, that brought forth the cultural practice and techniques of editing in the first place. The focus of the practice and the techniques was on stemming the progressive deterioration of the cultural heritage preserved in documents — a deterioration affecting not merely the material substratum of papyrus, parchment or paper (or, indeed, stone) as well as the inks and colours of inscription, but the writing, the texts themselves, in the documents. Hence, editing became predicated increasingly on the assumption that its cultural (as well as, in time, its scholarly) function was to produce restored and, emphatically, faultless texts, texts that could be adjudicated ‘right’, with all ‘wrong’ unambiguously recognised and eliminated; and texts also that could be claimed as ‘pure’. The juridical premise should give us pause, as should a fortiori the ethical one, so perceptibly tinged with religious morals.

29Such social and moral demands and expectations to bring textual scholarship and editing back, and perhaps down, again to their core and to the first premise of criticism, may contribute to altering significantly the basic assumption about the scholarly edition. It is not in its nature to produce and deliver faultless, that is, correct and pure, texts. This is so not only because the correct, pure, definitive (or however else adjectivally idealised) text simply cannot be achieved through any editorial practice, let alone theory; it is so also because to set such goals for editing is fundamentally misconceived, for they go against the nature as well as the historicity of texts and, by extension, the epistemological dimension of the voice through which texts are articulated, that is: the voice of human language that lives through the empowering energy of semantic multivalence.

  • 11 Gunter Martens, the German editor and editorial theorist, thoughtfully discussed the fundamental d (...)

30It is due to the always perceptible energy surplus in language that texts never come to rest — their simultaneous striving towards closure notwithstanding. Closure, admittedly, must be a real tendency of texts, one that also meets a readerly desire — or we could not have been living for so long in the happy delusion of being able, editorially or discursively, to attain definitive texts and interpretations. In truth, though, the critic as well as the textual critic and editor live alike in a constant see-saw alternation between a Newtonian and an Einsteinian understanding of the nature of texts.11

31It follows from the generals as well as the particulars of such considerations that a scholarly edition does not fulfil itself in the setting-forth of a text. A scholarly edition consists essentially, rather, as we have said, in the argument that holds together and unites its several orders of discourse. This proposition is of considerable importance for outlining a fresh perspective. The model I am advocating for scholarly editing in the twenty-first century is predicated on the functional correlation of bodies of material content in a systemics of discourses and argument. This model offers distinct advantages for mapping rationales and techniques of the digital medium for our new century’s endeavours to rethink (while enhancing the essence of) the scholarly edition as a product and instrument of learning, knowledge and professional skill.

32As such, the scholarly edition is at its core naturally an instrument of criticism. So to think, or re-think it means to link back to the traditions of longest standing within the historical endeavour to which scholarly editing belongs, and which are main traditions also in the cultural pedigree of editions. Scholarly editions of old, as we know, stood comprehensively in the service of understanding the texts and works they assumed their role in transmitting. Not only did their editors analyse, assess and evaluate textual transmissions as textual transmissions; they explored historically, critically and culturally, too, the works they presented in their editions. Negotiating between horizons of everyday life and sensibilities, of knowledge, thought and wisdom at a work’s times of origin as against its present-day moment of reception, such editions endeavoured to be comprehensive tools of mediation. The annotations and commentary with which they accomplished their acts of mediation were content-directed and contributed to generating an edition’s overall argument. To reestablish comprehensively the critical dimension of the scholarly edition, it is important to bring these traditions fully-fledged into focus again. Indeed, this is an essential precondition if and when we wish to project the scholarly edition of the twenty-first century as an edition capable of meeting the expectations of being based in criticism, scholarship and learning, and of being deployable, in its turn, as a platform and an instrument to enable critical analysis and generate knowledge and learning.

33The greatest opportunity, consequently, for innovation in scholarly editing as criticism through digital editions may ultimately lie in the field of the commentary. In crossing the divide from print to the digital medium, the scholarly edition is undergoing significant conceptual modifications. These will stand out most clearly through its commentary discourse. The edition as scholarly product stands the chance of becoming, that is: of becoming again, a node and main juncture of text and knowledge. Given that the task of ascertaining ‘the text’ is, and remains, a central concern of scholarly editing, it is at the same time also true that, more largely, editing is concerned, as it also always has been, with the historicity of texts. It is thereby implicated in the nature and historicity of human knowledge and understanding in which texts and their historicity are always already embedded.

34It is here, moreover, that the great opportunity of the digital edition lies in the relationality of the new medium. In the orthodoxies of annotation and commentary in the material medium of paper and the book, the embedding of the texts in their historicities has traditionally been almost exclusively understood as keyable, item by item, to the words and phrases of the text. The knowledge that textual and explanatory notes jointly carry has been seen as positive knowledge — which is why it has been feasible, and considered unproblematic, to chop it up into fragmented apparatus entries.

35Clearly, positive knowledge is not to be dismissed. It often enough makes sense simply to give a dictionary definition of a word, or to cite snippets of encyclopaedic factual information — as often as not, it may be all that we want to be told in the moment of reading. Nonetheless, we have in our day increasing problems with the stance of positivism behind such compilation, and item-by-item imparting, of positive knowledge. As both text and commentary editors with a sense of the bearings of their tasks will acknowledge, it takes systematic and comprehensive acquaintance with given areas of knowledge (say, history, politics, folklore, myth, daily life, social custom, law and the judiciary, religious, literary, philosophical or scientific contexts) to identify the meanings, implications and resonances of the words of the text and, in terms of a text’s diachrony, of authorial changes in revision — let alone to outline (say, in editorial introductions) identifiable structures of composition against which the meaning and significance of a text and work might subsequently prove interpretable.

  • 12 See, in much detail, the penultimate essay in this collection, ‘Argument into Design: Editions as (...)

36It is only on the surface, therefore, that knowledge appears to be positivistic and hence to lend itself to segmentation into consecutive items of positive information. Fundamentally, knowledge is hermeneutically relational. In terms of an edition of a work established from the heritage of documents testifying to it, moreover, knowledge is always contextual — with ‘contextuality’ being not its secondary, but its primary quality. For it is only by its contexts that a text is definable, and defined, in its words, meanings and implications: in what it says and does not say. It is to a text as a web of knowledge encoded in the specificities of the text’s language that an edition’s commentary answers, and it is thus quite specifically in the antiphonal responses of the discourses of text and commentary that an edition may be conceived of, and constructed, as a knowledge site. This, in itself, is not a new insight. In the Renaissance, when books first became the medium for editions, printers devised breathtaking layouts (adapted, in turn, from medieval manuscripts) for surrounding texts with commentaries, often themselves again cross-referenced. In effect, they attempted to construct in print the relationality of what today are called hypertexts. From such exuberant beginnings, conventions of marginalia and footnotes, as well as great varieties of indexes, have solidly survived. They have, in books, all been attempts to offset by relational cross-patternings the ineluctable linearity and sequentiality of the page and the codex.12

37But within the book to establish this third, relational, dimension against its material two-dimensionality has always been a rudimentary gesture, and has always depended on involving and stimulating the reader’s imagination and memory. For editions existing digitally, by contrast, the relational dimension is a given of the medium, and complex relationalities may be encoded into the digital infrastructure itself — indeed, they must be so encoded, since in the digital medium, the reader’s memory ceases to function as a constitutive factor of orientation in the way it does when we exercise our culturally acquired skill of reading books. This, consequently, is precisely where and when the relational commentary comes into its own. Constructed over a web of links, it should be envisaged as the modelling of the scholarly edition of text-and-commentary as a knowledge site. With its commentary discourse digitally systematised on such terms, the scholarly edition as the digital edition of the future may, in the last resort, be constituted as that hub of criticism and knowledge we would desire: to reflect, in multifaceted relationality, the given texts, works, and writers, their everyday reality, and the thought and worldview of their times. Scholarly editing of the future has the potential to distill, as well as engender, both historical study and criticism. Herein lies both the task and the vision for the digital edition in the twenty-first century.


1 There is a conceptual distinction, and thus a terminological one, to be made between ‘text’ and ‘work’. Briefly: if the tenet be that a ‘work’ is an abstraction projected from one or more material texts in which it manifests itself, then this has consequences, too, for conceptualising and situating scholarly editions. These may be editions of single texts; more commonly, though, they are indeed editions of the work, in that they correlate the work’s several manifest, genetically and historically distinct texts. Jerome J. McGann correlates the terms on similar lines in The Textual Condition (Princeton: Princeton University Press, 1991).

2 Time and again, to be sure, authors have themselves acted as editors of their works, and oeuvres. This is culturally a fascinating strand in the traditions of editing. Authors’ self-editions, however, are not generally scholarly editions.

3 The ‘knowledge site’ as a term has been given currency by Peter L. Shillingsburg in From Gutenberg to Google (Cambridge: Cambridge University Press, 2007).

4 See Alois Pichler, ‘Towards the New Bergen Electronic Edition’, in Wittgenstein After His Nachlass (History of Analytical Philosophy Series), ed. by Nuno Venturinha (Basingstoke: Palgrave Macmillan, 2010), pp. 157–72; and see also

5 Available at

6 What has been realised for Søren Kierkegaards Skrifter was, at the time of writing the present essay, very much in preparation only for the Norwegian flagship edition of Henrik Ibsens Skrifter. Some ideas as to how its future digital edition might be structured, anticipating tenets of the present essay, are contained in my review article of 2007, published in Norwegian as ‘Henrik Ibsens Skrifter under utgivelse’, Nytt Norsk Tidskrift, 24/4 (2007), 350–64. The original text in English is ‘Henrik Ibsens Skrifter in Progress’, in Editionen in der Kritik 3 (Berliner Beiträge zur Editionswissenschaft, vol. 8), ed. by Hans-Gert Roloff (Berlin: Weidler Verlag, 2009), pp. 292–311.

7 Back in 2002, at the annual conference of the Association of Literary and Linguistic Computing held in Tübingen that year, I voiced for the first time my contention that it was time we rethought the core discourses of the scholarly edition. The talk was video-recorded and is still available at In the matter of the commentary, I adduced the example of a Munich dissertation by Ulrike Wolfrum, ‘Beschreibung der Reiß’ — Festschrift zur Brautfahrt Friedrichs V. von der Pfalz nach London (1613). Entwicklung eines editorischen Modells für das elektronische Medium (München: Herbert Utz Verlag, 2006). The dissertation’s solution to its task of fashioning a commentary for the digital edition involved a rethinking of the notion of commentary in modular terms with regard to content, and in nodal terms with regard to digital organisation. Its main device was to plant index nest-eggs (as one might call them), that is, to define content modules and to set these as relational nodes of the editor’s, hence of the edition’s, own devising — in other words, to model the language-articulated substance of texts into ‘ontologies’ and from these to explore texts semantically. Virtualised commentary, so organised, moves, and allows an edition’s users to move, beyond the retrieval of flat information, and into exploratory modes of scholarship — that is, into deploying the edition platform both as a knowledge site and as foundation for critical interpretation of a text’s webs of meaning.

8 The standard workaround practice I happen to be aware of that has meanwhile established itself for XML markup as recommended through the Text Encoding Initiative (TEI) is to designate page divisions as ‘milestones’.

9 It was the experience of digitally editing draft manuscripts in particular that led me to propose a reversal of the hierarchy of terms in editorial scholarship in ‘The Primacy of the Document in Editing’, in Ecdotica, 4 (2007), 197–207. (The essay also appeared in French as: ‘La prééminence du document dans l’édition’, in De l’hypertexte au manuscrit. L’apport et les limites du numérique pour l’édition et la valorisation de manuscrits littéraires modernes (Recherches & Travaux, n. 72), ed. by Françoise Leriche et Cécile Maynard (Grenoble: ELLUG, 2008), pp. 39–51.) An earlier stage of reflection is to be found in ‘Textkritikens uttydningskonst’, in Filologi og hermeneutikk (Nordisk Nettverk for Edisjonsfilologer: Skrifter 7), ed. by Odd Einar Haugen, Christian Janss and Tone Modalsli (Oslo: Solum Forlag, 2007), pp. 57–80 (in Swedish).

10 Their faultlines in theory notwithstanding, the glory of the twentieth-century bibliographical way and copy-text editing in editorial scholarship was the ‘application of thought to textual criticism’ (according to A. E. Housman’s famous essay of 1921, ‘The Application of Thought to Textual Criticism’, Proceedings of the Classical Association, 18 (1922 [Aug. 1921]), 67–84), in terms both of logic of procedure and of critical discernment in the analysis of variants. Fredson Bowers, for one, in his textual edition of Tom Jones (Henry Fielding, The History of Tom Jones, A Foundling, ed. by Martin C. Battestin and Fredson Bowers (Oxford: Clarendon Press, 1974), performed a logical separation of text and document when grafting changes made for the fourth edition in a no-longer extant copy of the third edition onto the material substratum (his copy-text) provided by the first edition, and trusted his critical powers when assessing what textual differences were actually Fielding’s own revisions. The resulting edition text, needless to say, was thoroughly the editor’s constructed text.

11 Gunter Martens, the German editor and editorial theorist, thoughtfully discussed the fundamental double nature of texts as simultaneously striving towards closure while always remaining potentially open in an essay in German, ‘Was ist — aus editorischer Sicht — ein Text? Überlegungen zur Bestimmung eines Zentralbegriffs der Editionsphilologie’, in Zu Werk und Text. Beiträge zur Textologie, ed. by S. Scheibe and C. Laufer (Berlin: Akademie-Verlag, 1991), pp. 135–56.

12 See, in much detail, the penultimate essay in this collection, ‘Argument into Design: Editions as a Sub-Species of the Printed Book’, p. 315.


Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search