Where is the editor?
Resistance in the creation of an electronic critical edition
Printed scholarly editions of any type suffer, for intrinsic and external reasons, from the lack of being incremental and re-usable, and fail in presenting both the results of the historical-critical research and the archive on which the research has been caried out in such a way that it is of use to literary and textual scholarship. The electronic paradigm has, despite its enourmeous storage capacities and intrinsic re-usability, not changed anything, but has on the contrary established the illusion that both the "objective" archive and the "subjective" edition could at the same time be presented in one product, be it called an electronic archive or an electronic edition.
In this article I suggest a model for electronic scholarly editing that unlinks the Archival Function (i.e. the preservation of the literary artifact in its historical form and the historical-critical research) from the Museum Function (the presentation by an editor of the physical appearance and/or the contents of the literary artifact in a documentary, aesthetic, sociological, authorial or bibliographical contextualization). The digital archive should be the place for the first function, showing a relative objectivity, or a documented subjectivity in its internal organization and encoding. The Museum Function should work in an edition disregarding its external form displaying the explicit and expressed subjectivity and the formal orientation of the editor. The relationship between these two functions is hierarchical.
The implementation of the Archive/Museum model calls for a re-thinking and a re-orientation of the function of the editor.
Exhibitions are wonderful happenings. My visit to Making & Meaning Holbeins Ambassadors, an exhibition held in the National Gallery in London, 1 proved this once again to be true. Celebrating the 500th anniversary of Hans Holbein the Youngers (ca. 1497-1543) birth, this exhibition detailed the succesful cleaning and restoration of this most complex painting from 1533, and presented it in its newly-found splendour to the public. In eight different rooms, the visitor could not only look at the painting, but also learn a lot about Holbeins painting techniques and his painting practice including the several stages and kinds of drawings he may have needed in preparation of the actual painting. Further, there were rooms devoted to and focussing on the personal and political background to the creation of the picture (e.g. the Reformation), on the identification and biography of Jean de Dinteville and Georges de Selve, whose portraits are featured in the painting, on the identification and possible meanings of the books, the astronomical and musical instruments on the two shelves, on the contemporary fashionable technique of the anamorphosis which Holbein used to paint the skull in the foreground, and on other full-length portraits by Holbein. An easy-reference map prevented the visitor from getting lost and laid out the path the creators of the exhibition had intended to follow. Acoustiguide recorded tours were available and tired visitors or let us be more positive and call them the more interested visitors could sit down and watch the projection of a video. But the greatest experience was that I could stroll at my own pace through the exhibition, visiting the rooms I was most interested in first, checking everything I learned against the actual painting, and coming back to agree or not with the information provided, knowing that what I read and saw was the result of profound scholarly research.
Why is it then that I have never had a similar experience when consulting a scholarly edition of a literary work, although the exhibition referred to and a scholarly edition are both a product of the study of and the interference with the perceivable state of a work of art, intended to preserve and rescue it as an artifact. 2 Probably, I will argue, because the editor tries to combine the functions of the archive and the museum in one and the same scholarly edition of a certain type. 3
The German, and thus the Dutch, school of Editionswissenschaft traditionally distinguishes three types of editions, i.e. the historical-critical edition, the study edition, and the reading edition. Although each one of them is intended for a different audience with different expectations as to the contents and composition of the edition, they all should in theory be based on the same full historical-critical research needed to constitute a Zitierbare text. All three of them present the reader with a reading-text, but they differ in broad lines in the extensiveness of the commentary section and in the inclusion of an apparatus criticus or variorum. The smaller the intended public, the more extensive the report on the full historical-critical research will be. This research analyses the transmission of the text historically and comments on its genesis and history.
In the production process, the three types of editions stand in a hierarchical relationship to one another: a reading edition and a study edition can best be extracted from respectively a study edition and a historical-critical edition. But when putting this scheme to practice, the uncomfortable funding and publishing reality distorts this seemingly ideal and logical line which in theory runs from scholarly research to book-production. The high cost-factor of historical-critical research and the fact that these three types of editions are seldom published next to each other, dictates that this hierarchic line is seldomly respected: a reading edition e.g. is rarely based on historical-critical research. The financial reality of third parties forces the editor to make or accept a choice for one type of edition with a limitation and product-orientation of the research as a result or not to make an edition at all.
To overcome the impracticality of the rigid theory, the Swiss theoretician Hans Zeller proposed a sort of pick n mix model which he calls the Baukastenprinzip. 4 He sees the several parts of an edition as separable modules which are linked to each other, but can exist apart from each other. In practice, this means that a clear and edited text is published in one book volume (A) and further single volumes contain the apparatus (B), the commentary section (C) and further essays (D). By being incremental and easily updatable, this scheme has the great advantage of being publication friendly: the publication of the different volumes can be tuned to the publishers or the projects financial situation and depending on the type of edition wanted for publication, the editor only prepares the volumes which are needed according to the scheme. Another type of edition can easily be produced by the publication of the missing volumes, and the revision and publication of only one of them can result in a completely updated edition. So, Zeller not only provided a model for the production of affordable editions, but also found a way to circumvent the costly and labour-intensive enterprise of making a complete new edition.
Contrasting with this first set of editions, German editorial scholarship also distinguishes a second set which is archival in its research, results, and purposes. This archival set comprises two more types of editions, i.e. the facsimile and the archive edition, 5 the latter of which presents diplomatic transcriptions of all the documentary sources. Whereas these two types are more likely to be published in combination with each other, editions from the first and the second set seldomly appear on the market together, although they are in their contents complementary to each other. Roughly speaking, the interested scholar either meets with a representation of the witnesses in an edition of the second type or has to accept the supplied display of the editors editorial decisions conforming to his theoretical vision, together with the resulting edited text in an edition of the first type, without the opportunity to check the editors work against the documentary sources or witnesses. This lack of information can impose a limitation on literary scholarship based on editions of either kind. Since one of the basic mottos of editorial scholarship is: Never trust the author, never trust the publisher and never trust your own eyes, why then should the user of an edition have to trust the editor? Of course, it goes without saying that scholars in preparing their editions do a lot of research and acquire thus a real expertise in the writings they edit, 6 but since editing always means human interference guided by a certain theory or theoretical paradigm, it cannot be done objectively. Consequently, it is the users indisputable right to doubt the editors paradigm and question his subjectivity on the basis of detailed knowledge of the editorial decisions made in the constitution of the edition. Therefore, and to contribute to a meaningful be it technical debate on the text and its genesis, the user of a scholarly edition must at least be provided with the possibility to check everything against a presentation of the original documents. Up to now, a printed scholarly edition does, by its form and formality, not seem to fulfill these requirements.
The very form of a printed scholarly edition is intrinsically even more problematic in that it does not carry in itself any potential energy. A hard-copy edition is merely the result of the transformation of kinetic energy into one fixed form. 7 This has two consequences:
To sum up: even the most brightly created scholarly edition in print suffers, for intrinsic and external reasons, from the lack of being incremental and re-usable, and fails in presenting both the results of the research and the archive on which the research has been carried out in such a way that it is of use to literary and textual scholarship.
By emphasizing the disadvantages and the problematic position of printed scholarly editions I might have given the impression that the solution for the future of scholarly editing could be found in electronic editions. It is indeed true that Roger Laufer turned to electronic editing because his own critical apparatus of his edition of A. -R. Le Sages Diable boiteux had become even for himself completely illegible and it is true as well that an electronic edition carries in itself a lot more potential energy compared to its printed counterpart, but the optimism about hyperediting and the theoretically unlimited storage capacity of digital systems does not solve the main problem this paper wants to address, i.e. the seemingly practical impossibility to combine a critical and an archival edition in one product. Therefore I want to suggest a model for electronic scholarly editing that unlinks the Archival Function from the Museum Function. By Archival Function I mean the preservation of the literary artifact in its historical form and the historical-critical research of a literary work. Museum Function I define as the presentation by an editor of the physical appearance and/or the contents of the literary artifact in a documentary, aesthetic, sociological, authorial or bibliographical 14 contextualization, intended for a specific public and published in a specific form and lay-out. The digital archive should be the place for the first function, showing a relative objectivity, or a documented subjectivity in its internal organization and encoding. The Museum Function should work in an edition disregarding its external form displaying the explicit and expressed subjectivity and the formal orientation of the editor. 15 The relationship between these two functions is hierarchical: there is no Museum Function without an Archival Function and an edition should always be based on a digital archive.
The digital archive must at least contain the digital facsimiles of the comparable physical archive in uncompressed format, produced in conformance with the international consensus on digital imaging for preservation, 16 together with the corresponding bibliographical description and the diplomatic transcription of the primary sources. All texts should be encoded and stored in a standardized format in order to be machine-readable and processable, interchangeable and accessible in time. The archive itself must be incremental and its contents usable for scholarly research, e.g. by the development and inclusion of an intelligent search and retrieval engine. Since the creation of such a digital archive is and will always be the work of man, it cannot be valueless 17 or objective. One may argue that the production of digital facsimiles is an objective and mechanical operation, but trying to make an objective transcription of primary sources is a commendable but implausable project. Therefore, the objective should be a documented subjectivity from a historical point of view. This can be achieved by encoding the material, making use of the SGML (Standard Generalized Markup Language) tag sets created by the TEI (Text Encoding Initiative). 18
The data kept in the digital archive can then be used and tweaked by the editor to design and compose his own edition. 19 In this sense, the Archive/Museum model sees the edition, irrespective of its publication form, as a spin-off product based on the condition of the electronic archive at a certain point in time. For the rest, it should be what the editor wants it to be.
By adopting this Archive/Museum model, I believe not only that room is being created for alternative editions, but that the productive process of such editions are being facilitated by the existence of the digital archive. Next to this, the bringing into practice of the model will in fact provide the editor or candidate-editor with the possibility to apply a Zellerian Baukastenprinzip to the creation of an electronic edition.
The first fully electronic edition project on modern Dutch literature, the electronic Streuvels project concentrates on the 1927 novel De Teleurgang van den Waterhoek by the Flemish author Stijn Streuvels (1871-1969). De Teleurgang van den Waterhoek (The Decline of the Waterhoek), tells about the dramatic consequences of industrial technology (i.e. the building of a bridge with the purpose to connect two provinces) for a rural community in a small Flemish village called den Waterhoek at the beginning of the twentieth century. In 1927 the novel was published in installments in the Dutch literary journal De Gids and in the same year it appeared in print both in Belgium and in the Netherlands as a separate novel. Twelve years after its first publication in print, the novel appeared in a second revised edition (tweede herziene druk) in 1939. Streuvels did revise the text of his novel a first time when preparing the text which had been published in De Gids for the first edition. But this time, he did it more drastically, only retaining 73.4% of the text of the first edition. Probably because the publisher wanted a shorter and hence a more marketable book, Streuvels decided to cut out especially those passages which the catholic critique had fulminated against in their reviews of the first edition. A major part of the novel deals with the relationship between Mira, an obstinate and voluptuous village girl and Maurice, the shy but promising engineer from the city, who is in charge of the bridge-building. In depicting this relationship, Streuvels had, according to the catholic critique, used a language which was too suggestive or sexually too explicit for that time. But it goes without saying that by leaving out these passages, sometimes as long as sixteen pages, the relationship between the two forementioned characters changes enormously. Nowadays the book is completely out of print, and one can buy in second hand bookshops only an edition based on this revised text which has been reprinted thirteen times up to now, with the fifteenth edition in 1985. 20
This is, of course, excellent material for an edition, but by exploring the possibilities of the Archive/Museum model, the electronic Streuvels project, funded by the Royal Academy of Dutch Language and Literature of Ghent, Belgium (Koninklijke Academie voor Nederlandse Taal- en Letterkunde), wants to escape the bookish limitatons of a traditional edition project, by building a digital archive from which several spin-off products can and will be extracted. The first such products will be a scholarly reading edition published in bookform and intended for a large audience, and a new sort of study edition published on CD-Rom and intended for the academic communitiy.
The digital archive will eventually comprise the complete electronic version of the physical archive kept in the Archive and Museum for the Flemish cultural life in Antwerp (Archief en Museum voor het Vlaamse Cultuurleven, AMVC), i.e. electronic facsimiles of all the witness sources, 21 together with their full transcriptions and bibliographical details encoded in TEI compliant SGML markup. The electronic study edition which will appear on CD-Rom, will include the fully searchable texts of the pre-publication published in De Gids, the first edition from 1927 and the second revised edition from 1939, the digital facsimiles of the manuscripts and typescripts, a glossary list, a (genetic) chapter on the production and the transmission of the work, including relevant correspondency between the author and his publisher, and a study on the reception of the work. Taking the (edited) text of the first edition as the orientation text, this hypertext edition will link the different versions of the text on the paragraph level in order to show the variant readings. By making use of the TEI tag sets for the production of the electronic study edition, the production of the reading edition, which will appear in book form, is being facilitated. The encoded files of the text of the first edition (in the original spelling) with its glossary list and the (genetic) chapter on the production and the transmission of the work will go to the publisher, accompanied by a chapter in which, by means of example, one or two passages of the text are studied and compared to one another in their separate versions. With a minimum of work, the publisher can then typeset the reading edition from the SGML encoded instances.
The study edition will be an explicit editorial re-organization of (part of) the digital archive by making use of hypertext, without the pretention to both construct a coherent display and collect every shred of evidence. And this is exactly the point where I locate the problems with editions/archives as Jerome McGanns The Rossetti Hypermedia Archive and Peter Robinsons The Wife of Baths Prologue, which look to combine both the Archival and the Museum Function on the same level of the product. By trying to minimize the presence of the editor in a so-called objective and neutral presentation and description of the material, the editor in fact reaffirms the need for his presence by burying the user with thousands or millions of empty hyperlinks. Robinson himself acknowledges the problems with his edition/archive and His proposed solution is to provide the reader with explanations of the significance of the parts, to provide frankly interpretive signposts, to express editorial opinions about the value and significance of the materials presented. But this only seems to solve the problem: it will in fact lead to a shift from the archival purpose of the edition to the presence of the editor in the archive. By using hypertext only on the display level, i.e. on the subjective and selective editorial level of the edition, for the creation of hypertext is by definition subjective and selective, and not in the organization of the digital archive, the Electronic Streuvels project not only hopes to provide an alternative for the limitations of a paper edition, it also wants to explore the possibilities and implications of electronic scholarly editorial work based on the Archive/Museum model and in doing so, the project wants to be of assistance in the development of a rationale of electronic scholarly editing.
The theory and practice of the creation of an electronic edition is being countered by resistance from the side of the hard-copy focussed German school of Editionswissenschaft. Seeing the products of electronic textual criticism as mainly undermining the past solid position of the editor, there are three important paradigm shifts to which resistance exists. First, the dogma of the solid text as presented by the editor is undermined and replaced by the tolerant acceptance of a text as a contextualized possibility. Second, the presented text, together with the apparatus, the commentary section etc., is not the edition but one of many possible editions, composed and constructed by the editor as a spin-off product of the underlying digital archive, in a specific form and lay-out, serving a specific purpose and intended for a specific audience. The underlying digital archive is complete at any stage of its existence but is at the same time meant to expand continuously. Third, the function of the editor needs rethinking from the perspective of the theory and practice of electronic scholarly editing which also means the need for new skills. The implementation of the Archive/Museum model in the new function of the editor results in a bifurcation of the job.
On the level of the Archival Function, he is a document and source manager with bibliographic, textuist and programming skills. He is responsible for the intellectually open organization of the archive through documentation of his decisions in a standardized way to guarantee the greatest possible usability, longevity, expandability and interchangeability. Since the argument goes that the editor of a work makes the best literary critic because of the acquired expertise on the subject, the argument can as well go that whoever builds the digital archive makes the best editor. In Belgium the structural organization of the physical archives and the funding policy of the Fund for Scientific Research of Flanders (FWO) towards projects on textual criticism leave the editor with no choice but to build the archive himself. Therefore, the function of the editor needs a re-orientation away from literary criticism to documentalism, analytical bibliography and textuism (and hence back to its roots).
On the level of the Museum Function, the editor is in his capacity of a textual critic a creator and a supplier of spin-off products which display a re-organization of and a coherent study based on the digital archive. This again needs programming and, very often, graphical designing skills.
With this new and more than full-time function of the editor, the question may not be: Where is the editor?, but Where is he now?
Edward Vanhoutte is Scientific Attaché at the Koninklijke Academie voor Nederlandse Taalen Letterkunde (Royal Academy of Dutch Language and Literature) in Ghent, Belgium and he is the coordinator of scholary editing and editorial policy in Flanders. He is currently preparing an SGML-TEI based electronic edition of De Teleurgang van den Waterhoek by Stijn Streuvels. He teaches on text encoding, hypertext and electronic publishing at the Dept. of Germanic Language (University of Antwerp). His research concentrates on the theory of text encoding in preparing scholary editions, and on the use of hypertext in teaching literature.
Bowers, Fredson. 1976. Transcription of Manuscripts: the Record
of Variants. in: Studies in Bibliography, 29 (1976), p. 212-264
© Edvard Vanhoutte 1999Return to Human IT 1/1999