The Complete works of Willem Frederik Hermans can be regarded as one of the largest editing projects in modern literature ever undertaken in the Low Countries. With the preparations for the edition started at the beginning of this century, the Hermans-project is, as any scholarly edition nowadays, heavily influenced by and dependent upon a range of developments in the digital humanities. Throughout the project, digital technologies are abundant: digital transcriptions are being produced by optical character recognition (OCR), the project makes an extensive use of collation tools (Collate, CollateX), and all textual data are stored in XML- and TEI-compliant forms. Although the project aims at delivering a twenty-four volume printed edition, the extensive use of digital technologies creates many new possibilities and challenges for digital accessibility. For a project that tries to serve the more generally interested reader as well as the academic community, questions on accessibility go beyond the idea of simply providing digital access to sometimes highly complex textual and contextual data. We should also carefully consider the diverse audiences we aim at: we want to get literary critics and scholars interested in the essentials of textual scholarship again, and at the same time develop new strategies for the dissemination of twentieth-century literary heritage among a possibly interested, but more and more divergent general audience. Therefore, when generating new derivatives out of our editorial research, we will have to make smart, strategic choices. In developing new digital products, we should try to collaborate with partners in the educational field and cultural heritage.

Editing the Complete works of Willem Frederik Hermans

Willem Frederik Hermans (1921-1995) is widely regarded as one of the most important writers in Dutch twentieth-century post-war literature (see for a recent critical essay on the works of Hermans Frans Ruiter and Wilbert Smulders (2013)). Soon after the Second World War, Hermans managed to get his first novels and short story collections published, some of which caused considerable uproar among fellow writers and critics. De tranen der acacia's (1949, The tears of the Acacia's, not yet translated), for example, was thought to be pornographic, and with the prepublication of the first chapter of his next novel Ik heb altijd gelijk (1951, I'm always right, not yet translated) Hermans was accused of offending the Catholic community in the Netherlands, for which he was brought to court in 1952, but later cleared of all charges. Hermans achieved his critical breakthrough with the novel De donkere kamer van Damokles (1958), and from his Nooit meer slapen (1966) on, he came to be one of the most popular and bestselling Dutch novelists. At that same time, Hermans' work gained more and more attention from literary scholars, who were by then highly influenced by New Criticism, focusing on intrinsic textual analysis—a way of looking at the literary text that was encouraged by Hermans himself, who proclaimed in an influential essay, first published in 1961, that in "classic" literary fiction every single detail of the story needs to have functionality within the narrative as a whole (Hermans 2008a). In addition to novels and short stories, Hermans wrote plays, poetry, and collections of essays. He also translated Wittgenstein's Tractatus logico-philosophicus, wrote a biography of the nineteenth-century Dutch writer Multatuli, and produced a scholarly edition of that author's famous novel Max Havelaar (1860). Hermans was a very prolific polemicist, and in that capacity he was a feared critic in the Netherlands for decades. With his Mandarijnen op zwavelzuur (1964, Mandarins on sulphuric acid), a collection of highly polemical essays, he ruthlessly passed judgement on the works of a whole generation of well-established writers. With some of his most known novels translated into French, Spanish, German, and English during the past fifteen years, Hermans' work is receiving more and more international acclaim. Recent English translations include Hermans (2006a), Hermans (2007), Hermans (2008b); German translations include Hermans (2001), Hermans (2002), Hermans (2003), Hermans (2005); French translations include Hermans (2006b), Hermans (2009a); Spanish translations include Hermans (2009b), Hermans (2010). In the French Newspaper Le Monde, Milan Kundera (2007a) (see for an English translation Kundera (2007b)) characterized The dark room of Damocles as an "improbably rich novel," and Ian Brunskill (2006) in his English review of Beyond sleep in The Times described that novel as "a welcome if belated introduction to an original and challenging voice in modern European literature."

Immediately after Hermans' death in 1995, plans were made for a multi-volume scholarly edition. The actual work for this edition started in 2000, and after a preparation of five years, in November 2005 the Huygens Institute for the History of the Netherlands, and its partners the Willem Frederik Hermans Estate and the Dutch literary publisher De Bezige Bij, published the first volume of the Volledige werken (Complete works). In the summer of 2013, with the twelfth volume published, the series was halfway complete. By the end of 2015, two new volumes will have been published: the sixth and last volume of Hermans' novels, comprising four very diverse novels that were originally published between 1987 and 1995, as well as another volume with essayistic works, containing the highly controversial Mandarijnen op zwavelzuur. Hermans' biography of Multatuli and a last volume of Hermans' short stories are scheduled for 2016. It is estimated that all twenty-four volumes, each between 800 to 1,000 pages on average, will have been published within the next five years. The edition is aimed both at literary scholars and the general reading public. In the first instance, the Volledige werken are presented as scholarly editions, with a focus on a clear reading text accompanied by a detailed editorial commentary on textual history, including critical and scholarly reception. By making extensive use of the enormously rich personal archive of Willem Frederik Hermans, the commentaries also entail many fascinating excerpts of the correspondence between Hermans and his publishers, his fellow writers, his literary friends, and his enemies. The complex textuality of Hermans' work is not represented in the edition itself, which has no textual apparatus, but the printed edition is accompanied by a dedicated website with much more in-depth information on all textual witnesses and the textual work that resulted in the edited text (see http://www.wfhermansvolledigewerken.nl; with the appearance of every new volume in the edition, the website is updated).

As with other recent scholarly editions of modern literature by the Huygens Institute, such as the Volledige werken of Louis Couperus and the Volledig werk of Willem Elsschot, the Hermans edition is firmly rooted in Continental, i.e. German editorial traditions (Couperus 1990-1996; see for a detailed explanation on editorial principles Reijnders et al. (1992) and Elsschot (2001-2006); see for more detailed analyses on Continental editorial theory for example Eggert (2009) and Lernout (2013)). For each text in the edition, one single text version is chosen upon which the text for the edition is established. Both the Couperus and Elsschot editions were based on first printings of the works, but the works of Hermans are presented as an "ultima manus" edition, taking the last authorized published versions of the work as a starting point, and with good reasons: during his whole writing life, Hermans was extremely concerned with the production of his texts, which he kept changing over the years and over the scores of printings. Hermans himself wanted only the latest editions of his works to be available to readers. In an essay written in 1972, he stated, "I wish that all previous editions of books which have been reprinted in an improved version would crumble to dust as if by magic, even if the change only involves a comma" (H[ermans] 1972, 12). With Hermans, changes do not just involve commas, nor do they happen in a linear way. When changing words, sentences, or paragraphs Hermans sometimes went back years and years to the very first printed version of a text. One of the reasons for Hermans' constant revising of his texts was the need he felt to keep his work up-to-date for younger generations. In addition, he was always eager to strengthen the composition and especially the plot of his novels. Hermans, himself aware of his own on-going hurry and indestructible sloppiness during the creative process of writing, acted as a schoolish perfectionist as soon as the work was published: every single remaining slip of the pen, orthographic mistake, or flaw in the narrative that had been overlooked had to be expelled from the next printing as soon as it was discovered by Hermans. With all that authorial editing and revision taking place over a number of years, it is important to acknowledge that throughout the different versions of Hermans' works, there are many places in which authorial revision intermingles with press corrections (for example in orthography and punctuation) caused by printers, publishers, or readers informing the publisher or writer of assumed flaws. Since non-authorial changes play a significant part in the process of textual transmission, and in many cases cannot even be distinguished from the authorial, these textual changes are represented in the edited text (see for more socialized approaches towards critical editing for example McGann (1991) and McKenzie (1999)).

Although the Volledige werken of Willem Frederik Hermans are presented as a printed edition, the project is designed as a digital project. In order to establish the critical texts for the edition, the collation of all textual witnesses is pivotal. In the case of Willem Frederik Hermans, the textual material to be collated leads to a total of over 50,000 book pages. Since collating them manually would take far too much time, computer-assisted text comparison is necessary, for which, at an earlier phase of the project, Peter Robinson's COLLATE was chosen. In order to be able to add editorial information to the collated witnesses and to use the collation output as a basis for the digital copy to be sent to the printer, the collation output was converted into XML-TEI files, consisting of an edited base text with an inline variant apparatus including all published textual witnesses (for more details on the creation of the digital XML-TEI working environment out of the collation-output, or complexities relating to OCR and token recognition, see Kegel and Van Elsacker (2006)). Since 2005, the year the first volume of the edition was published, this textual database has served as the most important foundation for establishing the texts for the edition, and this tool is crucial in the workflow of publishing two new volumes of the Volledige werken each year. Throughout the years, therefore, much effort has been spent in taking care of technological sustainability. From 2013 on, the gradually more and more outmoded Collate software, running on an outdated Macintosh operating system, has been replaced by its successor CollateX, which has been developed within the European COST-Action programme Interedition. In preparing the texts for the edition, the XML-TEI working environment showed its richness and potential for editorial research by enabling the editors to analyse all text versions in a neat and concise digital representation, with the added advantage of using specific TEI-encodings for detailed accounts of editorial emendations and other editorial comments—for example, orthographic and stylistic idiosyncrasies—, thus greatly facilitating the research on the textual history of Hermans' novels. For De tranen der acacia's, for example, published in 2005 in the first volume of the Volledige werken, this textual history proved to be far more complex than was expected. While Hermans himself had always stated that his early novels had been written in a rather straightforward way, without taking much care to plan and construct the narrative, it turned out that he actually spent much time and energy in proofreading and correcting this novel before its first printing in 1949. Revised editions appeared in 1950 and 1953, and the process of revision and rewriting continued in later years. In preparing another revised edition, which was to be published in 1971, Hermans incorporated previous changes and alterations, such as modernization of style and orthography as well as more thematic and narrative reinforcements. Hermans kept revising the novel for many years to come, with the last corrected version being published in 1993, two years before his death (see for more details on the textual history of De tranen der acacia's Gielkens and Kegel (2008)).

Scholarly editing, digital textuality and digital society

While analyzing the textual variants within the XML-TEI working environment that was created for editorial research, it became more and more clear that this editing tool could also be useful as a starting point for creating a digital edition of one or more texts by Hermans. Digital access to the different text versions of Hermans' works could have a considerable influence on the way that work would be viewed. The mere possibility of studying the textual complexity in such a digital edition would break up the notion of a final and fixed textual structure, with many new opportunities arising for representing a wide range of textual changes that are all, each from its own perspective, relevant in the textual process. This way of looking at the literary work accords with John Bryant's The fluid text: A theory of revision and editing for book and screen (2002), in which he proclaims the notion of the work as something fixed and stable, and by consequence also an edition with "definitive" texts, as outmoded. Directly on the first pages of his book Bryant writes, "The very nature of writing, the creative process, and shifting intentionality, as well as the powerful social forces that occasion translation, adaptation, and censorship among readers – in short, the facts of revision, publication, and reception – urge us to recognize that the only 'definitive text' is a multiplicity of texts, or rather, the fluid text" (2002, 2). Already in 1996 Shillingsburg stated that "[the] word definitive should be banished from editorial discussion" (1996, 90). A more recent Dutch discussion on the subject is provided in Van Raemdonck (2009). In a digital edition based on this view, the diverse but interrelated manifestations of the literary work could be demonstrated with the different text versions functioning as "versions of each other" and new meanings and interpretations arising from the textual variation (Bryant 2002, 87). Though this textual condition has by now become widely accepted within textual scholarship, until this day little progress seems to have been made in convincing literary critics and scholars, let alone the general readership, of the potential benefits of this point of view. Yet to scholars, students, and probably even the more generally-interested reading public, these new digital representations of the work can present new and inspiring ways of discussing and analyzing literature. In the case of Willem Frederik Hermans, interesting examples are abundant. For instance, a literary scholar interested in the primary reception of a novel like Uit talloos veel miljoenen (981, Out of many millions, not yet translated) or De donkere kamer van Damokles could in a digital environment easily resort to the first original editions of these novels and compare them to the extended versions of later years, both of which were expanded with an extra chapter that considerably changed the plot of the narrative, a major textual revision that has attracted too-little critical attention.

But what should such a digital edition look like? How ambitious can such a digital project be, and for what audiences should such a digital application be built? In From Gutenberg to Google: Electronic representations of literary texts (2006), Peter Shillingsburg writes extensively on the possibilities and challenges for textual scholarship in the digital age, focusing on a conceptual framework for "knowledge sites," which he describes, in their ideal but yet unrealized forms, as "sites of textual complexity from which the beauty of complex coherence shines: where text and countertext, annotation and image, singularity and multiplicity of perspectives can serve readers upon whom nothing is lost" (Shillingsburg 2006, 24). Shillingsburg sets out the preconditions for such a digital edition and formulates guidelines for selecting which documents to include, methodological principles to be followed, relevant contexts to be taken into account, and usability questions to be dealt with. He pays special attention to frequently underestimated considerations in the design of digital editions, such as questions of legacy, project funding, and the development of a proper interface. He especially emphasizes the necessity of providing contextual material in the edition in order for literary critics and other audiences to take good notice of them, thus aiming to place textual scholarship within the centre of literary criticism again. Hans Walter Gabler comes to the same conclusion, envisioning the digital edition as a "hub of criticism and knowledge" by which the scholarly editing of the future "has the potential of distilling, as well as engendering, both historical study and criticism" (Gabler 2010, 54).

However sane those ideas may be and however relevant they are in the case of an author like Willem Frederik Hermans, they are only realizable for large and long-lasting international projects with a broad community of involved scholars and a potentially large readership as well. For projects operating in minor languages and/or small countries, such as the Huygens Institute's projects on modern Dutch and Flemish literature, those objectives tend to be far too ambitious, as is stressed by Edward Vanhoutte in an essay on the historical development and functionality of digital editions. Vanhoutte states rather pessimistically,

Editors of texts from such traditions work for an audience of only a few interested academics and a small reading public who for the most part want simply to read texts from printed books. 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 fantasy (2010, 120).

In two more recent, interrelated essays on the nature of literary works and their digital representations, Shillingsburg himself also tends to be less ambitious. In describing the scope and ambitions of digital editions, Shillingsburg argues for a "modular" organization of the digital material, with a variety of components that should be "connectable" and "extendable" (Shillingsburg 2009). With regard to the edited literary work, Shillingsburg explains that any literary work always should be considered as being "implied,""represented," and/or "interpreted," (Shillingsburg 2010) and that all interpretation of a work derives from interaction with an always partially represented work, meaning that "the only access we have to the work is through acts of interpretation of representations that imply the work. It follows that the work itself exists only in deferred forms—not in immediate, transparent, unambiguous forms" (Shillingsburg 2010, 179). In short, the work implied stresses the fact that a work in itself is always an abstraction, based on the act of reading a single version of a text. Since text versions imply the work differently, Shillingsburg emphasizes that versions in juxtaposition "imply the work in a more complex way" (2010, 172). A represented work always is a partial and particular representation, based upon one (or in a digital environment: more than one) text version, and in its representation cannot be the work itself; a digital representation then is just another partial and particular manifestation, which therefore implies the work in yet another way. In the same vein, but from the broader perspective of cultural heritage, Alan Galey asserts that any digital application developed within that domain will, with its new representation of the artefact, change the way it is being looked upon. In this respect much attention should also be paid to the development of a proper interface design, which in itself should be a core challenge for textual scholarship (Galey 2010).

In the process of developing new digital editions or other digital applications for the divergent audiences that a project like the Volledige werken of Hermans has to serve, there are other developments to be aware of. We should acknowledge the fact that the general reading culture is changing rapidly. More than ever before, the book trade seems to have become dependent on a few bestselling authors with huge overproduction in the market and an ever increasing rate of circulation, where there is no place left for the large majority of other writers and their books. This situation is just part of a revolution in reading culture where social changes go together with technological ones. During the last few years, we experienced, to quote Jim Collins from his Bring on the books for everybody,

a complete redefinition of what literary reading means within the heart of electronic culture. The really significant next new thing was not a matter of radical innovations in literary craft but massive infrastructural changes in literary culture that introduced a new set of players, locations, rituals, and use values for reading literary fiction (2010, 3).

Or, to frame this development in a broader context of book and digital media studies, as Adriaan van der Weel does in his Changing our textual minds: Towards a digital order of knowledge (2011), we are at the brink of a new world in which our "textual minds" are rapidly changing and the many book-based certainties of the homo typographicus are no longer obvious (see especially the final chapter 'Coda' in Changing our textual minds). This shift in the publishing landscape is extremely important. Since our Hermans edition is also a commercial enterprise and at least partly aimed at the general reading public, we should face the fact that in the very near future, if not today, at least part of our audiences will be far better served with other, and initially less or non-bookish forms of accessibility to Hermans' work. Given that situation, how do we reach a generation of readers who are, as described by Van der Weel (2011), more and more immersed in a digital space where textuality, though abundant, is one and maybe not the most prominent among other digital modalities, such as games, video, or music? How do we deal with a rapidly changing reading behaviour that increasingly consists of "zapping" texts like we "zap" songs and television programs and non-linear reading, of dealing with fragmentation in a chaotic and overwhelming textual and non-textual space, through which a personal path is taken, possibly in close interaction with the virtual community of the social Web? How do we make accessible our cultural inheritance "to the digital generation in a form that speaks to them?" (as described by Van der Weel (2011), 218; see also Van der Weel's chapter 1 in this volume).

Complete works and beyond: Explorations in accessibility

For us as editors of the Complete works of Willem Frederik Hermans, the tasks and ambitions we are faced with are manifold. In a time of scarce funding and an ever-faster changing digital textuality, we have to rethink our aims and ambitions in order to bring the merits of textual scholarship to the attention of the academic community, and at the same time find new ways to reach different targeted audiences within the general reading public. During the past years we have developed some small-scaled, tailored examples of new forms of accessibility for different user-groups. Our explorations, all aimed at conveying the specialized knowledge of textual scholarship, make full use of the challenges of digital accessibility, while we also try to take advantage of recent, not strictly digital developments within the broader field of editorial theory.

The XML-TEI working environment that was already created within the project served as an ideal basis for a first initiative to provide literary scholars with new forms of accessibility. In order to model the desirable characteristics and features of a TEI-based digital edition, in 2008 we developed a pilot digital edition of Hermans' short story "Paranoia," which originally appeared in 1948 in a small Dutch literary magazine and was part of the story collection Paranoia published by Hermans in 1953. The technical framework for the edition was provided by eLaborate, the online publication platform designed at the Huygens Institute. eLaborate is an online work environment in which scholars can upload scans, transcribe and annotate text, and publish the results as on online text edition which is freely available to all users. The publication platform of eLaborate can also be used for publishing XML-TEI encoded editions. In this pilot edition, which merely focuses on the presentation of the textual data, all digital text versions can be consulted, with accompanying images of the printed pages. In addition, detailed information on the textual history is provided, with in-depth descriptions of all relevant witnesses. The most salient feature of this edition is the possibility of comparing all textual witnesses by accessing an "inline" variant edition, in which any desirable combination of witnesses can be read against each other while being generated "on the fly" from the XML-TEI data (Kegel and Ravenek 2009). Although the default view is that of the edited text, users can employ any other witness as their reference version and can choose which other text versions they want to see represented alongside the reference version. Further, all known authorial corrections and revisions can be visualized, and this textual data is accompanied with detailed commentary on editorial emendations; on textual passages that seem flawed but nonetheless have good reasons for being represented in the edition; and on authorial revisions that are thought to be especially interesting for scholarly interpretation. Due to copyright issues, a digital edition based on the prototype is not yet available to the public, but the actual implementation of such a digital edition of at least one of Hermans' novels is expected in the future, after bringing the technology and interface in line with the latest technology.

While this pilot digital edition at its core focuses on the academic community, we also experimented with altogether new forms of accessibility for a younger, but otherwise vaguely defined audience within the general reading public. During the summer of 2011, the Huygens Institute and the Hermans Estate took part in an IBM Extreme Blue Project, IBM's top internship programme for business and technical students, which focuses on delivering innovative new solutions to inspiring challenges that face their clients. The challenge set out for four IBM trainees was: "How to ensure the heritage of Willem Frederik Hermans through modern technology for future generations." The students worked on a business plan for the solution and developed the prototype Beyond Read, an application that makes full use of the opportunities brought about by the emerging infrastructure of social media. Based on the data-mining of social media profiles coupled with an index of the works of Hermans created with IBM Content Analytics (ICA), readers received personal advice on which novel, short story, or essay by Hermans to read. For example, a Facebook profile showing interest in travel, walking, and Scandinavia, would lead to the suggestion to read (and buy) Beyond sleep. Based on this advice, Beyond Read created a new, personalized social reading environment in which the advantages of social media are combined with tailored, high quality data stemming from the editorial research for the Complete works. This thought-provoking experiment brought about new prospects for dissemination of the Hermans edition and the editorial research material behind it. The Beyond Read application was greeted with enthusiasm by the publisher De Bezige Bij and attracted attention from other parties involved in literary pedagogy and reading behaviour, such as the CPNB (Collective Promotion of the Dutch Book); Stichting Lezen (The Dutch Reading Foundation), whose mandate is to bring reading to the attention of a wide public; and Stichting Lezen voor de Lijst (a foundation supporting the teaching of reading for secondary school students), which has developed a coherent classification system for literary fiction based on students' individual reading experience. Based on the actual reading experience, the dedicated website http://www.lezenvoordelijst.nl offers secondary students a substantial amount of in-depth assignments for literary texts aimed at providing them with a frame of reference for the development of their literary competence. See http://www.literaryframework.eu/home.html for a European initiative based on the Dutch website.

Together with the Hermans Estate and other partners, a project proposal was written in which the IBM prototype was integrated as a teaser for a digital environment that would include most of the essential textual and contextual material, plus interactions and user enhancements, as formulated in Shillingsburg's (2006) knowledge sites. However, the realization of such a plan, as long as it was to be implemented in application to the whole project, would need considerable funding. Since it turned out to be much too expensive, with the added disadvantage that it would take too much time to realize such an ambitious project, we eventually refrained from applying for funding.

In the process of our reflecting on new forms of accessibility for the works of Willem Frederik Hermans, another important incentive was brought about by a one-day brainstorming session with a few leading Dutch creative agencies. This session, which took place in spring 2012, was organized to encourage unconventional thinking on the dissemination of literary heritage, and it led to concrete proposals for gamification of essential features of Hermans' writing (one of them being the immersive experience of enjoying an intricate literary narrative), to be communicated with an appropriate, clearly defined audience: culturally interested young people aged 20-25 and up with little leisure time but a strong desire for high-quality cultural and literary experiences. In our case, gamification procedures are at first instance focused on providing new forms of accessibility and are for example aimed at visualizing narrative (polemic) strategies employed by the author. For a discussion on the usability of game paradigms within the context of the social edition, on which much research is being done by the INKE Modelling and Prototyping Group, see for example Sapach and Saklofske (2013). The brainstorming session convinced us again that instead of aiming at overly ambitious projects that cannot get proper funding, we should focus on small, tailored solutions, providing new forms of accessibility for clearly defined user groups and developing them by collaborating with partners outside the academic world.

In 2012, we realized our ambition to engage a new and younger reading public with the literary work of Hermans. Hermans' seminal novel De donkere kamer van Damokles presented excellent opportunities for communicating some of the essentials of textual scholarship. In close coordination with the Hermans Estate and the Huygens Institute, the CPNB selected De donkere kamer van Damokles as the book of the year in the annual national campaign "Nederland leest" (The Netherlands read), the Dutch equivalent of the American "One Book, One City" campaign that originated in Chicago in 2001. From the beginning to the end of November 2012, nationwide reading and discussion of the De donkere kamer van Damokles were promoted: free copies of the novel could be obtained at every Dutch public library and were handed out in secondary schools, and throughout the Netherlands many public events were organized. The number of books printed for this occasion amounted to three-quarters of a million copies, for which the text of our scholarly edition was used, which in itself could be seen as something quite unique: never before in The Netherlands did a text based on a scholarly edition achieve such a wide circulation. Furthermore, during "Nederland leest" considerable effort was invested by the Huygens Institute in communicating some core characteristics of the novel from a textual scholarship perspective. In the midst of the campaign, the editorial work on the Volledige werken received attention from Dutch national television and one of the major Dutch national newspapers. In cooperation with the Dutch Reading Foundation, a short movie was made on Hermans' habit of continuous textual revision and was incorporated in their dedicated secondary school website for the campaign. See http://www.nederlandleest.nl/school/ for the annual campaign website. The short movie on Hermans can still be found at http://www.wfhermansvolledigewerken.nl/?p=921. For the more particularly interested general reader, an exhibition on the textual history of the novel was organized in close coordination with the Museum of Dutch Literature, where it was hosted. In all this, special attention was given to the first manuscript version of the novel, which was only discovered at the beginning of this century. This first sketch of what would become a large novel only during an intensive process of rewriting is highly interesting, and not only for the general public.

Especially for a scholarly audience, the manuscript version of the novel can be used to illustrate Hermans' poetics. De donkere kamer van Damokles, like the majority of Hermans' writings, is first and foremost an experiment with literary genre and narrative mode. Though the novel is one of the most interpreted literary works of twentieth-century Dutch literature, its experimental basis is largely overlooked in the critical reflection on the novel, which is too often read within an uncomplicated realist aesthetic. A clear focus on the textual history of the novel, based on a textual-genetic approach to the manuscript, can show what was really at stake as Hermans elaborated the text.

French genetic criticism focuses first and foremost on the creative process as it can be distilled out of manuscripts and other textual versions of a published work. On the basis of a critically constructed "genetic dossier" or "avant-texte," the creative process can be studied from a variety of critical perspectives, among them psychoanalysis, semiotics, linguistics and narratology (see De Biasi (2004), pages 42-43). Within this genetic approach, special attention is often paid to the opening ("incipit") and closure ("excipit") of the literary work. For an example of a genetic analysis of the "excipit" see Genette (2004). For a collection of essays on the "incipit" see Boie and Ferrer (1993). A genetic analysis of the "incipit" of the manuscript version of what eventually became the novel De donkere kamer van Damokles is especially relevant and revealing from a narratological point of view (see for a recent essay on the mutual benefits of interdisciplinary research in narratology and genetic criticism Bernaerts and Van Hulle (2013)). This first, unfinished version of the narrative, dating from 1952, shows that it was actually developed as a short story, with an unreliable first-person narrator as its protagonist. The story was originally meant for the story-collection Paranoia (1953), which as a whole can be regarded as an extensive exploration of the highly modernist topos of unreliable narration. And while the manuscript shows us how Hermans set up the narrative with a first-person point of view, it also makes clear how the story was directly based upon an historical case. The very first lines of the manuscript consist of a direct, nearly literal recapitulation of a well-known public case: the trial of a notorious Dutch double agent, who had, during most of the Second World War, been working at the same time for both the German occupiers and the allied forces. His life and adventures had been extensively written about in the national press during the late forties and early fifties, and Hermans used these historical particulars as the starting point for his story.

But the "incipit" is especially relevant for what happened afterwards. In reading the "incipit" of the story against the introductory chapters of the eventual novel, one becomes aware of the "textual invention" (Deppman, Ferrer, and Groden 2004, 11) that took place in the writing process. Reading the text versions of manuscript and novel against each other shows fully the extent to which the later version is, to quote genetic critic Pierre-Marc de Biasi,

The result of a process, that is, a progressive transformation, an investment of time that the author has devoted to researching documents, writing, correcting and re-correcting, etc. The literary work, closed in its perfected form and in a state of equilibrium that seems to be the immediate expression of its own internal necessity, nonetheless remains the mediated product of its own genesis (2004, 37).

This "progressive transformation" in the case of the Hermans' story was nothing less than a complete reconceptualization of the narrative, which took place more than two years after the creation of the manuscript version. Letters from the writer to his publisher at the start of 1954 show how Hermans, at that moment, had been totally reworking the story. By that time, the narration of the manuscript was preceded by a new opening of the narrative, indicating how to read the sequences of events that were to appear, with the earlier story being recomposed as a "small novel," as Hermans indicated his novel in progress in a letter to his publisher Geert van Oorschot, written November 30, 1953 (Hermans 2004). In revealing this revised design to his publisher, Hermans described his forthcoming novel as "highly sensational" with the role of his main character as "a highly bloodthirsty and sometimes questionable one" (Hermans 2004, 108). It would take Hermans another four years to finish the book, which had in spring 1958 grown into a novel as thick as a fist. In the published version of the novel, the revised opening of the narrative had grown into five short chapters. The explicit references to the historical case so prominent in the "incipit" of the story were removed, and the first person narration had been replaced by an only seemingly third-person narrative, behind which a narrator's voice now and then breaks through. Especially in these introductory chapters, written in an excessive, melodramatic mode, the occurrence of this narrator is patently obvious, providing the reader with a great many essentials on the main character Henri Osewoudt, thereby indicating how to read the narrative that develops throughout the novel. This melodramatic "mode of excess" (see Brooks (1976) for a detailed account on the history of the genre and the manifestations of melodramatic features in modern literature), although less overtly used in most other parts of the novel, nevertheless remains evident throughout, with a range of melodramatic topoi such as the overemphasizing of postures, gestures, and scent; descriptions of abnormalities, especially in facial expressions; a portrayal that is more focused on apparent traits instead of psychological complexity; and an ongoing play with notions of innocence and virtue (Brooks 1976).

Designing the near future

The importance of the melodramatic incentive, so often neglected in interpretations of De donkere kamer van Damokles, can be much better appreciated from a genetic perspective, taking into account the first manuscript version of the literary work. Elaborating the technical framework that we already created with the "Paranoia" digital pilot edition, a next challenge would be to integrate a digital transcription of the manuscript into a digital edition of the novel, based on the XML-TEI-files of the printed versions that are already available. New technological solutions provided by CollateX, among them the ability to detect textual transpositions (see Dekker et al. (2014) 452-470, especially "1.3 Comparing texts with CollateX," 457-461), can facilitate computer-generated collations revealing the complex textuality of De donkere kamer van Damokles within a digital environment. With such an edition of Hermans' seminal novel, we could finally present the scholarly community with a tool suitable "to map out the variation, chart paths from one version to another, and enable users to lead themselves along those paths" (Bryant 2002, 123).

However, as our other explorations described above have shown, with the ever-faster changing technology in which we are immersed, other digital devices and platforms should be used to reach the different, targeted audiences we aim at. To achieve this goal, we should build on our experiences with the IBM-project and creative agencies and develop small, modular projects focusing on a single text or specific feature of Hermans' work. Two recent initiatives may serve as an illustration of this ambition. First, together with the National Dutch Literary Museum, an online exhibition and educational platform for literary heritage— Literatuurmuseum.nl—is being developed. This cooperation between the Huygens Institute and the Literary Museum has a twofold ambition: for The Huygens Institute, as the main participating academic partner, the joint operation provides an opportunity to present the outcomes of academic research within the heart of literary education: for the literary museum, the collaboration with an academic partner opens up new ways of exhibiting literary resources by integrating current academic research. One of the digital exhibitions that will be developed shortly focuses on the first manuscript draft of Hermans' De donkere kamer van Damokles, thereby making use of the new possibilities for recontextualization and interpretation of the novel resulting from the genetic research. Second, we intensified our cooperation with Stichting Lezen voor de Lijst. While preparing the volume of our Complete works edition with Hermans' polemical essays Mandarijnen op zwavelzuur, we developed a reading dossier with detailed assignments based on this collection of essays, which has already been integrated into the foundations' national website for literature education (the assignments for Mandarijnen op zwavelzuur have been developed by Hester Meuleman). In this way, along with the publication of the volume within our Complete works edition, we simultaneously provide the practical means to incorporate this complicated collection of essays into secondary school teaching, thus directly addressing a new generation of readers.

With an author from a small country, writing in a minor language with a relatively small and diverse interested audience, one should capitalize on what comes to hand, by providing well-considered examples that provoke interest. As long as it "takes familiar objects and puts them in a new and liberating context," (Robinson 2010, 152) a new, tailored form of accessibility could be digital as well as non-digital. In developing new initiatives, close cooperation with other players in the field of literature can help us as textual scholars achieve our goals: making the literary work of Willem Frederik Hermans productive for new explorations and interpretations .