Being Practical. Electronic editions of Flemish literary texts in an international perspective

This is the text of my lecture at the International Workshop on Electronic Editing (9-11 February 2012) in the School of Cultural Texts and Records at Jadavpur University, Kolkata, India.

The slides of this lecture have been published on Slideshare.

Keep it cool: the electronic edition & the fridge

Over the last couple of years, I have been observing my children's continuous development of skills with growing amusement. And, as those amongst you who are parents or grandparents will agree, kids sometimes really amaze you. From the age of two, my boys know, for instance, how to operate a fridge. As far as I can recall, neither their mother nor their father taught them how it worked and I'm pretty sure the grandparents didn't tutor them privately either. Nevertheless, they have since been very successful in opening the door of the fridge, exploring (if not rummaging) the contents, finding what they are craving for, picking one or two incidentally found extra's on their way out, and running off with their treasures after having closed the door again. They also noticed quite early on that the light is operated automatically on opening and closing the door and that they don't need to use a switch for that.

While I was witnessing one of their recent scavenger's hunts, it occurred to me that the fridge was the perfect model for what we have been looking for for over almost two decades now in the design of electronic textual editions. A fridge is an intuitively designed repository of a diverse range of foods from which anyone may quarry what they need. It offers an ideal storage space for a selection of fresh meat, fish, vegetables, fruit, dairy products as well as for semi-prepared foods, finished dishes, and leftovers. It is also the most economic and safest option to defrost foods. Although there is a generally acknowledged plan by which a fridge should be filled – bottles go in the inside of the door, vegetables and herbs go in the boxes at the bottom, meat goes on the bottom shelf and dairy products go on the top shelf – the internal organisation of the foods on the shelves is decided on by whoever fills it up, and can be changed according to the insights and preferences of any user. The products can, for instance, be grouped according to food group, meal, user frequency, size and so on. The fridge can be refilled, products can be replaced by fresher ones and new products can be introduced. Another feature is that one only needs some pieces of paper and a couple of magnets or post-it notes to annotate the contents of the fridge, put up shopping lists, or leave instructions about the next meal. By the same technique the appearance of the fridge is altered on a daily basis by moving around the notes, introducing new ones, taking old ones off, embellishing the outside with various collections of fridge magnets or with your children's artistic creations. The fridge's main function is to preserve foods over a certain period of time and to offer easy access to a wide range of products from within people's homes. And fridges are available in many models with various features like freezing compartments, ice makers and water dispensers which extend the fridge's central function. Unfortunately, it must be admitted, a fridge can't cook you a meal.

Electronic editions, by comparison, or at least the electronic editions we want to be making, should be intuitively designed 'repositories of information, from which skilled scholars might quarry what they need' as Peter Robinson stipulated once (Robinson, 2003b). Michael Sperberg-McQueen reminded us that 'any edition records a selection from the observable and the recoverable portions' of an 'infinite set of facts to the work being edited.' (Sperberg- McQueen, 2002) He mentions the apparatus of variants, glosses for some words, historical or literary annotations and the like as possible selections and 'visual effects, atmospheric sound, music, film clips of readings or performances' as possible elements of inclusion. Scholars who have written on the preferable contents of electronic editions like Susan Hockey (Hockey, 1996, p. 13-14), Marilyn Deegan and Peter Robinson (Deegan and Robinson, 1994 [1990], p. 36), Peter Shillingsburg (1996b) and Thomas Tanselle (Tanselle, 1995b, p. 592) agreed already in the mid-nineteen nineties that full accurate transcriptions and full digital images of each witness were essential parts of the edition, and both Tanselle (Tanselle, 1995b, p. 591) and Shillingsburg (1996a, p. 95) added to this the requirement of critically reconstructed texts. Whereas the 1997 CSE Guidelines for Editors of Scholarly Editions were fairly prescriptive on the contents of an electronic edition,1 Dan O'Donnell observes in 2005 that no 'standardization exists for the electronic editor' (O'Donnel, 2005b) while he points to electronic editions without textual introductions, without critical texts, without traditional textual apparatus, and without glossaries. The last version of the Guidelines published in 2006 (CSE, 2006), however, does not prescribe the contents of an electronic edition anymore, but provides some generalizations of the methodological orientation by which specific materials are edited. These Guidelines state that reliability is a defining quality of the scholarly edition and that this can be established by accuracy, adequacy, appropriateness, consistency, and explicitness. About how this is achieved in the edition, the Guidelines only observe that 'most scholarly editions' include a general introduction and explanatory annotations; they 'generally' include some sort of editorial statement; and 'commonly' include documentation of alterations or variant readings in appropriate textual apparatus or notes. In the attached list of Guiding Questions for Vetters of Scholarly Editions to these Guidelines, one could find all elements of the contents of an electronic edition as prescribed in the 1997 version of the Guidelines, but no minimal or defining score for scholarly editions of any nature is given. Robinson (2007a, p.8) summarizes: 'for a digital edition to be all it can and should be, then it will let the editors include all that should be included, and say all that needs to be said.'

The use of a general markup language like the TEI for the encoding of the electronic edition's contents and its exploitation by publication suites that take advantage of this encoding allow the organisation and reorganisation, that is the grouping and selection, of the edition's data according to a variety of principles. One of the earliest noted advantages of the electronic edition is, as far as it is not published on a fixed medium like a CD-ROM, as was the case in its early days, its openness to revision and change. But how we should keep track of the subsequent versions of such an edition is another matter which we may discuss in the session on preservation on Friday. The issue of third party annotation creation and display in a digital edition is another much debated central component of the electronic edition we want to be making (e.g. Robinson, 2003a; Boot, 2007a; 2007b). As a matter of fact, user driven annotation tools were often integrated in the SGML publishing software with which early electronic editions were published (e.g. De Smedt & Vanhoutte, 2000) but disappeared when interfaces to electronic editions were built on the basis of open source engines and suites of XML related formatting, stylesheet and query languages. Possible models to empower the user have recently been proposed by Shillingsburg's knowledge sites (Shillingsburg, 2006) and Ray Siemens' social edition (Siemens et al., forthcoming). Also on Friday, we can discuss this further in the session led by Anna Gerber.

The main function of an edition, whether it is conceived of and published electronically or in print, is to mediate, as Paul Eggert has reminded us, 'according to defined or undefined standards or conventions, between the text of a document made by another and the audience of that anticipated publication.' (Eggert, 2002, p. 17) Thereby the editor is involved in taking attitudes towards the preservation, presentation, and transmission of an existing text (Eggert, 2002, p. 17-18). Consequently, the electronic edition must contain the data to present a text and ways to explicate the editor's attitudes. On top of that, the electronic edition may contain analytical tools by which the user can replicate the editor's methodology and data processing. Another essential function of the edition is what I am calling the communicative function, namely to make sure that it reaches as wide an audience as possible.

This is where the fridge-model does not represent the reality of electronic editions anymore. The relative failing of the electronic edition has been lamented by their creators on many occasions (Robinson, 2005; 2010; Vanhoutte, 2009). Overall, the existing electronic editions have lacked to find their audience and thus failed in their communicative function. As an undergraduate student taking my philosophy exam I had to answer the question whether a chair on Mars was useful. The correct answer, which I happened to produce, was that the usefulness of a chair on Mars was dependent on the presence of subjects to whom the chair could be useful. If we take for granted that Mars is not populated by subjects who could appreciate the functional qualities of the chair, that chair is useless and functionally non-existent. Peter Robinson once claimed that 'an edition is an act of communication.' Consequently '[i]f it does not communicate', he says, 'it is useless.' (Robinson, 2009 [1997-2002]) By contrast, the fridge appeals to everyone, from the food addict, the really hungry, and the professional chef, to the keen amateur cook, the incidental snacker and the complete novice, to the food hater. From the omnivore, to the health guru, to the vegetarian and the vegan. But if fridges and electronic editions have so much in common, why is it then that not every household has at least one of each? The fridge's success can be explained by its consistent offer of the same functions and opportunities to all human beings and through them even to animals – dog and cat food are also preserved in the fridge. On top of that, the basic interface and functions are fixed and independent of what colour, size, type, or design the fridge itself takes. The fridge's success is thanks to its design for one culture. The electronic's edition's failure is due to its design for two cultures.

The problem of two audiences and two natures

John Lavagnino identifies the communicative function of the edition as problematic and points at an anomaly in the context of academic publishing. Whereas scholars across all disciplines mainly publish within the circle of their peers and address a larger community in popularized writings, 'a scholarly editor', according to Lavagnino, 'is still always expected to serve a larger community that may not – and, at present, usually does not – take any great interest in the discipline of editing.' (Lavagnino (2009) [1997-2002]) In this greater community, which Lavagnino also names 'the popular audience' or 'the common reader', he includes many scholars who haven't had any involvement or interest in editing, and thus don't understand the codes of the scholarly edition. The tension between serving both the common reader and the editor's peer with the same product he calls 'the problem of two audiences'.

The literary critic is in the first place a reader, possibly an academic, and exceptionally a textual critic or a scholarly editor. As Dirk Van Hulle has reiterated, '[L]iterary critics tend to take the text for granted by assuming that the words on which they base their interpretations are an unproblematic starting point.' (Van Hulle, 2004, p. 2) Scholarly editing as a product generating activity can react to this observation in two extreme ways. The first possibility is not to contest the literary critics' or common readers' assumptions about the definite singularity of the text and provide them only with the result of scholarly editing, namely an established text, preferably accompanied by annotations of some sort. The second option is to confront them with their wrong assumption and draw their attention to the multiplicity of the fluid text caused by its genetic and transmissional history. This can be done by introducing them to the data of textual scholarship.

The first option is a function of the reading edition, which I am calling the minimal edition (Vanhoutte, 2010), the second option is a function of the historical-critical or variorum edition, which I am calling the maximal edition. The minimal edition is a cultural product that is produced by the scholarly editor who acts as a curator or guardian of the text, whereas the maximal edition is an academic product that is produced by the scholarly editor who demonstrates their scholarly accuracy and scrutiny. The minimal edition is targeted towards well but negatively defined audiences – that is, readers who are not interested in scholarly editing – and present only the conclusions of the full critical and historical research on the genetic and transmissional history of the text. Besides an editorial statement and some sort of commentary, they most importantly present a citeable text which can be enjoyed as an aesthetic reading object. The maximal edition, then, presents the critical and historical research itself in an attempt to engender understanding amongst the editor's peers.

The two audiences are neatly but separately served by the minimal and the maximal edition which are essentially different in nature. Therefore, the commercial reality of scholarly editions of the minimal and the maximal type should be taken into consideration when theorizing about their essential function and audiences.

But who are the other audience – the editor's peers? I guess there does exist a small but growing group of literary critics who do take some interest in scholarly editing. Besides this group, there are other scholarly editors who may be interested in a scholarly edition for a variety of reasons. One group may be interested because they are editing or have been editing the same text. Given the reality of scholarly editing, however, this is very unlikely, except, perhaps, for a couple of important and much debated texts. Another group may consist of editors who work or have been working on editions of texts by the same author. Often, these editors work in close collaboration with each other and hence don't really form the most critical group. A third group consists of editors who work or have been working on texts from the same period or the same literary tradition; texts with a similar document architecture or complexity; or texts with a similar transmissional or genetic history. This group is interested in the edition's theoretical solutions to a variety of problems offered by the text. A last group consists of editors who are interested in another editor's methodology, the edition's technology, and the editorial models suggested, explored, and demonstrated by the edition.

The first two groups are interested in all of the presented texts across the fullest appearance of the edition, including the record of variants, the commentary and all other parts of the edition that are focused on gaining an understanding of the text. It has to be added that their interest may be comparative in nature. The third group is only interested in those parts of the texts which are problematic to scholarly editing and useful to their own purposes. The interest is often in methodology, visualisation procedures, and techniques. The last group is not interested in the text or its meaning, but only in the technology applied to the text and the edition.

The members of all of these groups change according to the edition. Since I have mainly edited works by Flemish authors, for instance, I belong to the first two groups whenever an edition of works by one of these authors appears or in the very unlikely situation when a text I have edited is edited again. I belong to the third group, however, when consulting editions of works of Flemish authors I haven't worked on or of a modern foreign author. However, I most often belong to the fourth group when looking at editions of classical, mediaeval, or renaissance texts, or any electronic edition I can lay may hands on. A member of the first two groups, I will study the integral edition in the hope the edition will add to my understanding of the text. A member of the third group, I will try to isolate some samples in the edition which demonstrate a problem akin to the problems I'm working on in the hope that the edition will add to my understanding of how to deal with these kind of problems. My attention will go in particular to the editorial statement and the textual essay explaining the applied methodology. A member of the last group, I will explore the edition's functionalities and architecture, study the edition's markup and rendering, including its design, and try to understand the edition's technological issues. My attention will go in particular to the technical documentation, the encoding strategies, and the source files I can rip off the edition.

Being practical

At the Centre for Scholarly Editing and Document Studies of the Royal Academy of Dutch Language and Literature, we create editions for both of these audiences and exploit either the print or the digital medium. For the common reader, we're editing a series of complete works by modern Flemish poets which are published in print. Since 2004, four volumes appeared with the collected poetry of Jos De Haes (1920-1974), Hugues C. Pernath (1931-1975), Paul Snoek (1933-1981), and Eddy Van Vliet (1942-2002). These poets are roughly comparable in name and fame to British poets like Philip Hobsbaum, Philip Larkin, Stevie Smith, and John Silkin. For the kind of editions with an explicit cultural function, we developed the text-critical edition which presents itself explicitly as a reading edition, but contains elements which are traditionally found in a study edition, for instance concise annotations and the textual essay containing chapters on the genetic history of the text, on the transmission of the text and the bibliographic description of the extant witnesses, and on the editorial principles. The textual essay is written from the perspective of the reader who wants to be informed about the reading text rather than from the perspective of the textual scholar who wants to demonstrate the results of their research. It has to be noted that this textual essay can easily be ignored, which is why we print them after the reading texts. The books are easily between 500 and 900 pages, are published in paperback and are sold for no more than € 30. The edition of the complete poems of De Haes sold 585 copies in three years time, Pernath sold 663 copies in two years time, the edition of Snoek – the most voluminous one – sold 1,126 copies in only three months time, and Van Vliet sold 756 copies in three years. And it's not only poetry that sells. An annotated reading edition of the selected letters of Herman De Coninck (1944-1997), another Flemish poet and critic, sold over 3,800 copies in two years time. I'm convinced that the successful sales figures of these editions are thanks to the unambiguous focus on the common reader who wants to read texts as aesthetic and historical objects.

There is no use in taking advantage of digital technology for the publication of electronic editions of these collected poetry because the audience is simply not there, and the common reader who is buying the print edition does not want the electronic edition. In 1999-2000 I edited a novel by Stijn Streuvels. The text-critical reading edition was published in 1999 by a literary publisher and sold all of the 375 copies which were printed in less than three months time. Although the demand was there for a second print, the publisher was not interested and refused. A year later, the electronic-critical edition of the same novel, targeted at a more academic public was published with an academic publisher. Some 200 CD-ROMs got sold and I still consider this a success. It seemed however, that there was a clear distinction between the different audiences of either product. The reading edition was completely sold to the group of common readers. The electronic edition, however, was sold to a more diverse group of people, namely the four groups of peers I introduced above, accompanied by common readers with an interest in scholarly editing, collectors of Streuvelsabilia, policy makers, and the technologically curious. I estimate that the anticipated audience which was really interested in the genetic and transmissional history of the text as explained by the electronic edition, consisted of about thirty people. At least, I have the impression I know each one of them by name. But the momentum was there for the electronic edition. The publisher had just returned from a visit to Jerome McGann's Rossetti Project in Virginia when I contacted her. She had heard about SGML, and I offered her an SGML encoded edition at that very moment. After a 15-minute chat and a demonstration, the contract was signed for the production and distribution of 500 CD-ROM's. I know that I will never be able to close such a deal ever in my life again. In 2007 I managed the production of probably the most successful electronic edition on CD-ROM ever. A total of 2,350 copies of the electronic edition of Willem Elsschot's Achter de Schermen was produced. But none of them was sold individually: 650 copies were inserted into Dirk Van Hulle's book on genetic criticism, 750 copies were sold to the Elsschot Society who gave it out as part of the membership package, 960 copies were presented as a Xmas gift to the contacts of the Dutch Huygens Institute, and 90 were retained for marketing and demonstration purposes.

These figures are interesting, because they suggest that these electronic editions have found their audience and they argue against the failure of the electronic edition. However, their audience has come by accident to the electronic edition. These electronic editions were not successful because they were state of the art products of textual scholarship, but because of the immense popularity of the original author, because of the inclusion of digital images of the manuscripts, which always appeals to collectors, because of its novelty character, or because of the big give away campaign. These accidental audiences were never taken into consideration when producing the edition. But if this accidental audience does become the target audience of electronic editions that is instrumental in the fulfilment of the communicative function, the edition must also provide this audience with access to a text and access to understanding by means of the same product. If this is the case, we are in the business of creating electronic editions of two cultures.

Editions of two cultures

The electronic edition distorts the efficiency of this system by ignoring the problems of two audiences and two natures in trying to combine two cultures in one product. The technical possibilities of the electronic edition brought to scholarly editing the option of all-inclusiveness which led early anticipators like Shillingsburg to visions of blurred distinctive lines among electronic archives, scholarly editions, and tutorials (Shillingsburg, 1996b, p. 25). Three central qualities of the electronic edition answered the call in conventional scholarly editing for the discipline's movement towards a true science, namely storage capacity, text encoding, and visualization technology.

The cornerstone of true science is the principle of external replication. This means that the scientific results or data obtained under conditions which are the same each time should be reproducible by peers in order to be valid. Further, the report on the research should contain sufficient information to enable peers to assess observations and to evaluate intellectual processes (Council of Biology Editors, 1994). This is exactly what maximal editions do by the presentation of their formalized and formulized apparatuses – apart from providing the data for a more or less correct assessment of the genetic and transmissional history of the text. The scientific reflex in editorial theory could hence be interpreted as the recognition that the function of the maximal edition is not to inform the reader but to protect the editor. This is why I call these maximal editions ambiguous and ambidextrous. Ambiguous because the presentation of the genetic and transmissional variants subverts the stability of the reliable textual basis the literary critic is looking for, but at the same time, the presentation of an established reading text may be too speculative for geneticists and scholars interested in the variant stages of the work. Ambidextrous because a maximal edition logically contains a minimal edition and presents the textual archive alongside. The key feature of the electronic edition, then, in order to appeal to many audiences would be a differentiation of the supply by user controlled selection mechanisms which can turn the all-inclusive edition into a minimal version presenting one citeable text accompanied by selected categories of commentary. Only, as I argue elsewhere (Vanhoutte, 2009), the electronic edition, despite its dynamic architecture, fails through its medium as cultural product and can't compete with any printed version of the text which is easily available for the reader. We should have learned by now that common readers seldomly turn to the screen for aesthetic experiences other than those offered by the exposition of full colour digital facsimiles of exceptional manuscript material.

This was cleverly exploited a couple of years ago by the Dutch Royal Library which provided on-line access to the full colour facsimiles of the famous Flemish Gruuthuse manuscript shortly after their acquisition caused a political scandal in Flanders. The Flemish common reader on the one hand argued against their own government who had let the manuscript leave the country, but on the other hand praised the newly achieved access to the facsimile edition which was just right: an introduction, the digital facsimiles and a transcription offered in two interfaces, namely a Flash version which allows you to browse through the manuscript, and an HTML version which presents the digital facsimiles next to the transcriptions. Also very clever from a marketing point of view is the direct entrance to the most famous song in the manuscript offered from the welcoming page of the on-line exposition. This on-line edition provides access to data rather than understanding and is a huge success with the public thanks to its singular focus on one culture.

The central assumption of the electronic edition that the reader's understanding of a text is better catered for by a capacious edition representing a multitude of versions and states of the text under study and databases of critical analyses and commentaries submitted by the (co-)editors and critical user is based on the utopic concept of the professional student of the text, not on the concept of an educated and interested reader with other professional occupations. Electronic textual editions are highly specialized tools that are only understood by scholars who are akin with the principles and functions of textual editing and have read the users' manual. Editions for everyone are therefore a utopic concept.

Editions for everyone

In From Gutenberg to Google, Peter Shillingsburg introduces the concept of the Knowledge Site as an elaboration of his early vision of the blurring distinctive lines among electronic archives, scholarly editions, and tutorials (Shillingsburg, 1996b, p. 25).

The space and shape I will try to describe is one where textual archives serve as a base for scholarly editions which serve in tandem with every other sort of literary scholarship to create knowledge sites of current and developing scholarship that can also serve as pedagogical tools in an environment where each user can choose an entry way, select a congenial set of enabling contextual materials, and emerge with a personalized interactive form of the work (serving the place of the well-marked and dog-eared book), always able to plug back in for more information or different perspectives. (Shillingsburg, 2006, p. 88)

The knowledge site would provide the information needed to understand the meaning of textual variation rather than information needed to preference one text over another or separate right from wrong readings. Peter Robinson's concept of 'fluid, co-operative and distributed editions' (Robinson, 2003a, p. 125) that are truly actively interactive4 through their instinctive interface design (Robinson, 2007a; forthcoming a; b; c)5 realizes Shillingsburg's concept of knowledge sites through the formation of active on-line communities.6 This is a loud echo of Lavagnino's suggestion of a model for electronic editions based on interactive, collaborative work on texts: 'In this model, you not longer have the sharp division between producers and consumers of information [...] an interactive and collaborative edition would instead be open to incorporating work from everyone who's interested in contributing'. (Lavagnino, 1997-2002) Also, Robinson's ideas of 'electronic editions for everyone' (Robinson, forthcoming b) correspond with Shillingsburg's concepts of the convenient and the practical edition (Shillingsburg, 2005) that must bridge both the theoretical and practical differences between textual and literary critics and that goes back to Fredson Bowers' concept of the 'practical edition'.7

These new models of distributed and collaborative editions Shillingsburg and Robinson develop, however, will not provide the general model for electronic editions nor will they propose a generally applicable and stable interface for electronic editions that would approximate the fridge model. The distributed and open model for electronic editions may well be suited for the specific texts from classical, medieval, and Victorian Anglo-American textual traditions Robinson and Shillingsburg are involved with and they may well respond to the needs of the broad communities interested in them, but they may prove less useful for editors of texts from smaller and language specific traditions.8 Editors of modern Dutch and Flemish texts, for instance, work for a mostly receptive audience of only a few interested academics, and a reading public of a couple of hundreds who mainly want a practical reading edition in print. The idea of the active involvement of a computer literate and critical community with a knowledge site built around a modern Dutch or Flemish text is but an idle phantasy.

Add to this another characteristic of the average scholar which C.P. Snow already observed in his seminal 1959 Rede Lecture entitled The two cultures and the scientific revolution, namely that intellectuals are Luddites, further complicates the case of the distributed collaborative knowledge site-like edition. The theoretical model of the electronic edition for everyone as envisioned by Robinson, will in practice be the most specialized edition thinkable for the smallest group thinkable, consisting of editors of the same work or text and the same author and those literary critics interested in the scholarly edition of this specific work.

Being practical, again

Therefore, at the Centre for Scholarly Editing and Document Studies, we have developed a model that considers the electronic edition as a maximal edition that logically contains a minimal edition. An essential function of this maximal edition is that it fulfils the users' need for a reliable textual basis by the inclusion of a critically established reading text. Rather than providing a valuable supplement to a print edition, as is often the reduced function of an electronic edition in an editorial project, this model empowers the user to check upon the choices made in the critical establishment of the text by way of access to the textual archive. At the same time, the model allows the user to ignore the editors' suggestions and to develop their own perspective on the maximal edition, or to generate a minimal edition of their choice. The reproducability of the thus generated minimal edition is guaranteed by a record of the choices that informed it. This documentary feature of the electronic edition facilitates the scholarly debate on any one of the many texts and provides any reader with a clear statement on the status of the minimal edition generated and printed for distribution or reading. Because of the scholarly basis of the electronic edition as a whole, even the most plain reading text with no additional information generated by the user qualifies as a scholarly edition. By emphasizing the on the fly generation of user defined printable editions as a central feature in our system together with the documentation of its definition, we strive towards the re-evaluation of scholarly editions as cultural products. So we see the electronic edition – or the maximal edition – as the medium par excellence for the promotion of the scholarly reading edition – or the minimal edition – and the recentering of the printed edition.

I will demonstrate this with the electronic edition of De trein der traagheid which will be published online next May, after having served for many years as our tinkertoy for our experimental modeling approach. The edition currently presents a critically established reading text and nineteen versions of the novella from its print history. The result of the collation of all versions is documented according to the TEI parallel segmentation method inside a master XML file that also contains all editorial annotations. This guarantees the completely equal treatment of each version of the text in the generating processes invoked by the user. Through the interface of the edition, the user can exploit the underlying TEI encoding by selecting any version and generate three possible views of the texts: XML for analysis, XHTML for consultation on the screen, and PDF for printing out as a reading edition. Any version can also be combined with any combination of any number of witnesses whereby the initial version functions as orientation text and the other selected versions are displayed in a lemmatized apparatus variorum. From within this apparatus, the generated edition can be reoriented from the point of view of any included witness. The model applied to this specific textual history allows the user to generate 10,485,760 possible editions of the complete text of the novella and when taken into account that editions for each separate chapter can be generated as well, this figure is multiplied by 35 which gives a total of 367,001,600 possible editions.1 Any one of these editions can again be exported to XML, XHTML, or PDF. Any number of versions, depending on the dimension and resolution of the user's screen, can also be displayed in parallel and the respective lists of variants can be generated on the fly.2 The minimal and the maximal editions are fully searchable, and the search results can be displayed in a KWIC concordance format.

The edition is powered by a dedicated suite of open source XML-aware parsers, processors, and engines combined with appropriate XSLT, XQuery and XSLFO scripts.

Concluding remarks: Teach the audience how to swim

On the 1997 Toronto Conference on Editorial Problems, Michael Sperberg-McQueen and Peter Robinson wrapped the thesis of their papers in a swimming metaphor. Sperberg-McQueen advised the audience not to teach their edition how to swim, but instead concentrate on the content, not on the behaviour of the edition in order for it to survive. His paper summoned its audience to invest in the data and to use encoding standards like the TEI for that purpose. In a later revision of that paper, Sperberg-McQueen retained the metaphor but reversed its polarity, explaining now how to teach your edition how to swim. In that revision he refined his earlier focus on the content by adding that editions should also be given the capabilities 'of doing things interactively with the reader.' (Sperberg-McQueen, 2009 [1997-2002]) Peter Robinson, in his paper, replied to Sperberg-McQueen by contending that: '[T]he great promise of electronic editions [...] is not that we will find new ways of storing vast amounts of information. It is that we will find new ways of presenting this to readers, so that they may be better readers. To do this,' he added, 'we will have to teach our editions to swim to the readers.' (Robinson, 2009 [1997-2002]) Discussions on text-encoding, he called 'dry-land swimming' and in order to 'make some real editions for real readers' he reminded us why editors have to learn to swim.

Robinson's concepts of editions have been based on the anticipated reader which forms the essential basis for his understanding of text and meaning ('Text does not exist outside the meanings we create: and these meanings are all the text we will ever know.' (Robinson, 2009 [1997-2002])); for the purpose of encoding ('We do not ask: what is the right encoding of this word. We ask: who is to use the text we make? What use do they want to make of it? What do we think this text is saying? How can we, as editors, help the text speak to its readers?'); and for the edition ('A transcription, an edition, is 'right' only in that it might serve these purposes'). Or in a more direct formulation which appeals to Sperberg-McQueen's point: 'Editions do not survive because they are preserved in elegant encoding and in government-maintained electronic archives. They survive because they are read. They survive because people find them useful, they survive because scholars, students, school children find they help them read.' (Robinson, 2009 [1997-2002])

In my lecture today, I have tried to illustrate why I think this last statement is problematic when tested against the reality of electronic editions. Instead of teaching how (not) to teach your edition how to swim, or teaching the editors to swim, I suggest that we start teaching the audience to swim. When I, just minutes away, argued that electronic textual editions are highly specialized tools that are only understood by scholars who are akin with the principles and functions of textual editing and have read the users' manual, I meant what I was saying. We need to accompany our electronic editions with detailed manuals that outline the functionalities of the edition; that explain the anticipated audience how they can make use of the knowledge which has been put into the edition; how they can operate the included tools; how they can replicate the editor's research; how they can interact with the edition; and how they can contribute to the edition. We have to publish in papers and essays examples of the research questions generated by the edition as a trigger for scholars to dive into the edition and come up with suggestions and hypotheses which might solve them. We have to continue the communication about the discipline so that working with electronic editions becomes more academically acceptable.

But above all, as editors, we have to make sure, that each audience is allowed to swim in the pool which is best suited to their skills and purposes. I think we all know how annoying it is to try to float in an attempt to reach the state of weightlessness in a pool full of lane swimmers, and to try to swim lanes in a pool full of floaters. If the fridge was a utopic model for the electronic edition, a well organised swimming pool is a realistic one.

References

• Boot, Peter (2005). Advancing Digital Scholarship using EDITOR. In Humanities, Computers and Cultural heritage. Proceedings of the XVI international conference of the Association for History and Computing 14-17 September 2005. Amsterdam: Royal Netherlands Academy of Arts and Sciences, p. 43-48.
• Boot, Peter (2007a). A SANE approach to annotation in the digital edition. In Braungart, Georg, Gendolla, Peter and Jannidis, Fotis (eds), Jahrbuch für Computerphilologie, 8: 7-28. Also published in Jahrbuch für Computerphilologie - online.
• Boot, Peter (2007b). Mesotext. Framing and exploring annotations. In Stronks, Els and Boot, Peter (eds.), Learned Love. Proceedings of the Emblem Project Utrecht Conference on Dutch Love Emblems and the Internet (November 2006). The Hague: DANS, p. 211-225.
• Bowers, Fredson (1969). Practical Texts and Definitive Editions. In Hinman, Charlton and Bowers, Fredson, Two Lectures in Editing: Shakespeare and Hawthorne. s.l.: Ohio State University Press, p. 21-70.
• Council of Biology Editors (1994). Scientific style and format: the CBE manual for authors, editors, and publishers. 6th ed. Cambridge: Cambridge University Press.
• CSE (1997). Guidelines for Electronic Scholarly Editions. (1 December 1997, revised June 2002).
• CSE (2006). Guidelines for Editors of Scholarly Editions. (Last revised 7 April 2006) Also published in Burnard, Lou, O'Brien O'Keeffe, Katherine, and Unsworth, John (eds.) (2006). Electronic Textual Editing. New York: Modern Language Association of America, p. 23-46.
• Deegan, Marilyn and Robinson, Peter (1994). The Electronic Edition. In Scragg, D.G. and Szarmach, Paul E. (eds.), The Editing of Old English. Papers from the 1990 Manchester Conference. Cambridge: D.S. Brewer.
• De Smedt, Marcel & Vanhoutte, Edward (2000). Stijn Streuvels, De Teleurgang van den Waterhoek. Elektronisch-kritische editie/electronic-critical edition. Amsterdam: Amsterdam University Press/KANTL.
• Eggert, Paul (2002). The Importance of Scholarly Editing, and the Question of Standards. In Plachta, Bodo and Van Vliet, H.T.M. (eds), Perspectives of Scholarly Editing / Perspektiven der Textedition. Berlin: Wiedler Buchverlag.
• Lavagnino, John (2009 [1997-2002]). Access. In Julia Flanders, Peter Shillingsburg & Fred Unwalla (eds.) Computing the edition. Thematic Issue of LLC. The Journal of Digital Scholarship in the Humanities, 24/1: 63-76.
• O'Donnell, Daniel Paul (2005b). O Captain! My Captain! Using Technology to Guide Readers Through and Electronic Edition. The Heroic Age. A Journal of Early Medieval Northwestern Europe, 8.
• O'Reilly, Tim (2005). What is Web 2.0. Design Patterns and Business Models for the Next Generation of Software. O'Reilly. http://www.oreilly.com.
• Price, Kenneth M. (2007). Electronic Scholarly Editions. In Siemens, Ray and Schreibman, Susan (eds.), A Companion to Digital Literary Studies. Malden, MA / Oxford / Carlton: Blackwell Publishing, p. 434-450.
• Robinson, Peter (2003a). Where We Are with Electronic Scholarly Editions, and Where We Want to Be. In Braungart, Georg, Eibl, Karl and Jannidis, Fotis (eds.), Jahrbuch für Computerphilologie, 5: 125-146. Also published in Jahrbuch für Computerphilologie - online.
• Robinson, Peter (2003b). The History, Discoveries and Aims of the Canterbury Tales Project. The Chaucer Review, 38/2 (2003): 126-139.
• Robinson, Peter (2005). Current issues in making digital editions of medieval texts–or, do electronic scholarly editions have a future? Digital Medievalist,. 1.1 (Spring 2005).
• Robinson, Peter (2009 [1997-2002]). What text really is not, and why editors have to learn to swim. In Julia Flanders, Peter Shillingsburg & Fred Unwalla (eds.) Computing the edition. Thematic Issue of LLC. The Journal of Digital Scholarship in the Humanities, 24/1: 41-52.
• Robinson, Peter (2010). Electronic Editions for Everyone. In McCarty, Willard (ed.), Text and Genre in Reconstruction. Effects of Digitalization on Ideas, Behaviours, Products and Institutions. Cambridge: Open Book Publishers, p. 145-
• Shillingsburg, Peter L. (1996a). Scholarly Editing in the Computer Age. Theory and Practice. Third Edition. Ann Arbor: The University of Michigan Press.
• Shillingsburg, Peter (1996b). Principles for Electronic Archives, Scholarly Editions, and Tutorials. In Finneran, Richard J. (ed.), The Literary Text in the Digital Age. Ann Arbor: The University of Michigan Press, p. 23-35.
• Shillingsburg, Peter L. (2006). From Gutenberg to Google. Electronic Representations of Literary Texts. Cambridge: Cambridge University Press.
• Snow, C.P. (1961). The two cultures and the scientific revolution. London: Cambridge University Press. Originally delivered as the 1959 Rede Lecture.
• Siemens, Raymond, Timney, Meagan, Leitch, Cara, Koolen, Corina & Garnett, Alex (forthcoming 2012). Toward Modeling the Social Edition: An Approach to Understanding the Electronic Scholarly Edition in the Context of New and Emerging Social Media. Forthcoming in LLC. The Journal of Digital Scholarship in the Humanities.
• Sperberg-McQueen, C.M. (2009 [1997-2002]). How to Teach Your Edition How to Swim. In Julia Flanders, Peter Shillingsburg & Fred Unwalla (eds.) Computing the edition. Thematic Issue of LLC. The Journal of Digital Scholarship in the Humanities, 24/1: 27-39.
• Tanselle, G. Thomas (1995b). Critical Editions, Hypertexts, and Genetic Criticism. The Romanic Review, 86/3: 581-593.
• Vanhoutte, Edward (2009). Every Reader his own Bibliographer: an Absurdity? In Deegan, Marilyn and Sutherland, Kathryn (eds.), Text Editing, Print and the Digital World. Aldershot: Ashgate, 2009, p. 99-110.
• Van Hulle, Dirk (2004). Textual Awareness. A Genetic Study of Late Manuscripts by Joyce, Proust, and Mann. Ann Arbor: The University of Michigan Press.