Apl. Prof. Dr. Patrick Sahle arbeitet vor allem für die „Koordinierungsstelle Digital Humanities“ der Nordrhein-Westfälischen Akademie der Wissenschaften und der Künste und kümmert sich im Rahmen dieser Kooperation mit dem Cologne Center for eHumanities (CCeH) um DH-Belange der Akademie und ihrer Forschungsvorhaben. Außerdem ist er an der Leitung des CCeH beteiligt und koordiniert den Aufbau des „Data Center for the Humanities“ (DCH) an der Universität zu Köln.
Er ist u.a. Vizepräsident von QUANTUM, der – Association for Quantification and Methods in Historical and Social Research, Vorsitzender der DHd-AG „Datenzentren“, Teil des International Editorial Board von DH Commons, Mitglied des Unterausschusses „Geschichtswissenschaft in der Digitalen Welt“ im Verband der Historikerinnen und Historiker Deutschlands und Mitglied des Beirats zum Digitalen Archiv NRW.
Seine Dissertation zum Thema „Digitale Editionsformen“, die unter der Betreuung von Prof. Manfred Thaller im Fach Informationsverarbeitung entstanden ist, wurde im Mai ’09 an der Universität zu Köln eingereicht und liegt seit Anfang 2013 gedruckt und elektronisch vor.
In der Vergangenheit war Patrick Sahle für eine Reihe universitärer Einrichtungen sowie die Niedersächsische Staats- und Universitätsbibliothek Göttingen tätig. Zuletzt war er u.a. als wissenschaftlicher Mitarbeiter am Lehrstuhl für Historisch-Kulturwissenschaftliche Informationsverarbeitung für DARIAH-DE tätig oder als Lecturer an der Universität zu Köln an der Entwicklung und Durchführung eines Studienmoduls „IT-Zertifikat der Philosophischen Fakultät“ im Rahmen des „Studium Integrale“ beteiligt. Zu den größeren Projekten aus der entfernteren Vergangenheit gehörten unter anderem das e-Journal „HSR-Trans„, das Handschriftendigitalisierungsprojekt „CEEC“ oder die Portalprojekte „zvdd“ (digitalisierte Druckwerke) und „OPAL-Niedersachsen“ (digitalisierte Museumsobjekte). Hinzu kamen und kommen eine Reihe kleinerer Projekte, Auftragsarbeiten und Beratungstätigkeiten für Dritte.
Bereits seit seinem Studium der Geschichte, Philosophie und Politik an der Universität zu Köln (mit anschließendem Studienjahr an der Vatikanbibliothek in Rom) interessiert er sich vor allem für Fragen der digitalen Aufbereitung der Überlieferung und ist auf diesem Feld mit eigenen Projekten, Projektberatung, gedruckten und digitalen Publikationen und Lehrveranstaltungen aktiv. Seine Interessen erstrecken sich aber darüber hinaus auf den gesamten Bereich der „Digital Humanities“.
Für weitere Informationen siehe man seine ästhetisch und technisch immer noch den Geist des letzten Jahrtausends atmende Homepage
Introduction: Why do we need a definition?
1Humanities research is focused on cultural artefacts such as texts, images or physical objects. Usually they are kept in libraries, archives and museums and are thus not encountered as original material objects; rather, scholars work with surrogates of them created especially to make them more accessible and to facilitate research. Over the last centuries, the desire to uncover the cultural treasures of the past and to reconstitute important documents, texts and works in the most reliable way possible has led to the development of the concept of the critical edition in the modern sense. This implies the application of wide knowledge, ranging from material and bibliographic criticism to historical understanding and textual criticism, and can lead to very complex forms of publications. Editions are created by the best experts in a field. They establish reliable sources for research, authorise and canonise certain readings, and thus channel and frame our perception of history, literature, art, thinking, language etc.
2Accordingly, critical editing is a central field in the humanities, spanning nearly all disciplines and subjects. Over time, it has evolved into an independent research area offering a large corpus of theorising literature, sophisticated methodologies, learned associations and societies, dedicated conferences and journals and even specialised study programmes at the postgraduate level. In continental Europe at least, the discipline bears the distinct label Editorik or Editionswissenschaft in German and ecdotica, ecdotique or ecdotics in Italian, French and English respectively. Despite–or perhaps even because of–its relevance for many different subjects, there is no comprehensive definition for critical editions that would extend its validity beyond single genres, types of documents, transmission settings or methodological approaches. The idea of critical editing originates and has been most developed in literary studies of classical and medieval texts. Here, the attempt either to reconstruct a lost ‘original version’ (Urtext) from antiquity, or to realise the author’s will and intention for texts that have been contaminated and altered in the processes of transmission is central. Obviously, these ideas are bound to very specific settings of the creation and transmission of texts. They depend on particular theoretical assumptions and have been questioned in their goals ever since. The most prominent interpretation of critical editing as textual criticism thus seems rather narrow in a more global perspective.1
3Scholarly digital editions (or SDEs) offer the opportunity to overcome the limitations of print technology. The new possibilities have a fundamental impact on the theory and methodology of critical editing in general. The large corpus of research literature on digital scholarly editing is full of discussion about new features, functions, properties and characteristics of this kind of scholarly endeavour and how this changes our practices, our theoretical grounds and our goals. Among the pioneers of digital editing, Peter Robinson in particular has proposed a clear and condensed set of central aspects which come close to a definition.2 Yet, this erudite characterisation describes core properties rather than defining what a SDE actually is. When Robinson claims, for example, that a scholarly digital edition must ‘enrich reading’, then this names one of the most important goals of such an edition but does not yield sufficient criteria to distinguish the digital edition from other things.
4We still lack a clear definition to help us identify and label things as SDEs. So the question remains: how do you recognise a digital scholarly edition when you see one? As an important tool in our common methodology, we need a sharper knife here: what are we talking about when we talk about SDEs? What is part of the population and what not? Are we talking about the same things and the same set of objects? Or are we talking at cross-purposes? To develop our theories and methodologies further, we need to know more precisely where our empirical basis lies and what the population is. This has both theoretical and practical relevance. In my own work over the last twenty years I have used a working definition to collect and catalogue SDEs.3 This definition now seems sufficiently well tested to be ripe for presentation and further discussion. If it has been functional for classification and cataloguing, it can serve as a tool in further empirical and systematic research. And this, in turn, is essential in building the new methodology and theory of scholarly editing in the twenty-first century.
Premises and goals
5In the attempt to give a definition, we have to be clear about our scope. Digital editions are created in many different disciplines. Beyond the most formative impact of philology and literary studies, disciplines such as historical linguistics, history, art history, philosophy, musicology or archaeology have their own requirements and traditions. In the past, methods from philological disciplines have most often been adapted to other fields, although some have also developed their own approaches. Sophisticated editions require large amounts of time and money, making it unlikely that a subject, once dealt with, will be tackled again soon. Moreover, foundational work such as digital representation and basic transcriptions may be created only once. For this reason, editions should be as useful across disciplinary boundaries as possible. While this already suggests a common methodology, the demand for a shared definition increases with digital media and online editions that should be as widely accessible and usable as possible for all interested disciplines. The same holds true for the various textual genres, types of documents and materials. The notion of scholarly editing should not be restricted to literary texts but has to cover all cultural artefacts from the past that need critical examination in order to become useful sources for research in the humanities.4 Undoubtedly, the idea of critical editing according to the Lachmannian paradigm–that is the application of the highly developed techniques of textual criticism for the reconstruction of a lost Urtext or an author’s intention–is the most influential and most highly developed approach within the field. There have, however, always been dissenting approaches and other schools of thought.5 The construction of a reading text beyond the actual document witnesses as the main goal of editing, for example, has been questioned right from the start. In the study of history, textual criticism had been adopted but was flanked by historical criticism, accounting for the properties of the material documents as well as explaining the textual information by summarising, annotating or indexing them. For a broad understanding of the field, further schools such as critical bibliography, diplomatic or documentary editing, genetic editing or documentary editing have to be considered. None of these areas should be neglected in the attempt to come to a comprehensive definition. This may, in a first step, delineate what scholarly editions are, before we then move on to digital scholarly editions. In addition, building upon the editions that we know already should help us to keep the tradition and to save the scholarly achievements of the past. An adequate and productive definition should integrate all editorial schools, address all humanities disciplines, cover all textual genres and every kind of object. And, no less important, a good definition should be short and simple.
A simple definition for scholarly editions?
6As previously mentioned, in collecting and cataloguing scholarly editions to serve as an empirical basis for deeper analysis, I have been working with my own definition for many years now. This definition simply reads:
Edition ist die erschließende Wiedergabe historischer Dokumente.
7In German, this works quite well. Unfortunately, however, it relies on the central, yet untranslatable, term erschließen, which encompasses any activity that increases the amount of information concerning a specific object and thus enhances its accessibility and usability. Depending on context, words such as develop, open up, deduce or infer may be used to render this concept in English. They do not cover the wider notion intended here, however. To capture the basic idea that all of these processes involve making thoughtful, reflective and reasonable judgments about the objects of study, the word critical may not only be an approximation but an even better label for the concept. Thus, I propose the following definition:
A scholarly edition is the critical representation of historic documents.
8This surely fulfils the requirement for a short and simple statement, but it also contains four fundamental points that need to be discussed and explained further.
9First: representation. Representation means the recoding of a document or an abstract work and its transformation in the same or another kind of media. This is usually done on the visual layer by image reproduction or on the more abstract textual layer by transcription. Representation spans a wide scale between materially oriented reproductions of documents and the constitution of new readings of a text–e. g. in the attempt to reconstruct and realise a lost original or an author’s intention. Representations try to capture objects in their entirety and can be further transformed into publications. This already indicates a possible distinction between representation and presentation which will be discussed later. For the moment it is important that representation is a necessity for an edition. Critical engagement without representation is not an edition–but an examination, a catalogue or a description.
10Second: critical. To achieve a broad understanding of all kinds of scholarly editions, those that are out there now or which could possibly be created in the future, we have to start with an open and wide notion of critical. We cannot constrain our understanding to the most prominent exponent of criticism, which is textual criticism. Regarding editions, we have to take into account that there exist as well historic criticism, bibliographic criticism, material criticism, visual criticism and other forms of criticism. In short, criticism must stand for all processes that engage in a critical or reflective way that is, on the basis of a scholarly agenda–with the material in question and help in ‘opening it up’.
11Criticism as a practice and a process may take different forms. Think of the rules that are applied in the transcription of a document. While the transcription itself is a representation, the specification of rules and their application make this a critical process. Identifying structures, named entities or other objects of interest and making them explicit, e. g. by annotation, is yet another form of criticism. Judgments about punctuation, orthography, wording, corrections and emendations are typical tasks of textual criticism and are sometimes seen as the highest form of philology. In addition, free or formalised descriptions of the documents, their texts and their treatment within an edition are the backbones for the edition’s authority and reliability for its further scholarly usage. Finally, a critical attitude is required to decide with which additional material, to what extent and in what form an edited text should be contextualised in order to make it more understandable and accessible. To sum up, we may take the word critical as a container for all those activities that apply scholarly knowledge and reasoning to the process of reproducing documents and transforming a document or text into an edition. The critical handling of the material is a second necessary condition for an edition. A representation without such treatment or the addition of information is not an edition–but a facsimile, a reproduction or–nowadays–a digital archive or library. Critical representation as a compound notion of editing aims at the reconstruction and reproduction of texts and as such addresses their material and visual dimension as well as their abstract and intentional dimension.6
12Third: documents. Most editions focus on texts or even works. I prefer the term documents here for two reasons. On the one hand, not all editions have works or texts as their primary goal. Some material simply does not have textual content, or the textual content or the notion of an abstract work behind its physical embodiments is not central to the edition. At the same time, there are disciplines, schools and theoretical approaches where the material document itself lies at the heart of the editorial interest. However, every non-abstract object that is the subject of an edition can be called a document. On the other hand, text can be described as a function of documents.7 In the real world, the document is always the antecedent. Even if an edition is built upon an abstract notion of text or work, it always starts with material documents. Even if an edition tries to establish a certain text reading or version, beyond the evidence of textual witnesses reconstructing a lost Urtext or constituting from all witnesses as the best text a ‘text that never was’,8 this is based upon documentary evidence.9 Accordingly, document can–at least functionally–completely cover the notion of text.
13Fourth: historic. Editions are created for all disciplines. Historic does not mean for history here. Editions are the groundwork for further research and for reliable, authoritative texts that can be used in humanities scholarship and teaching. They explore the uncharted circumstances of documents, texts and their transmission. They may correct errors introduced by the conditions of production, copying and publishing. They explain what is not evident to the present-day reader. In short, they bridge a distance in time, a historical difference. Texts that are created today do not need to be critically edited. They can speak for themselves. Only historic documents and texts need an editor to make them speak clearly.
What is a scholarly digital edition?
14A scholarly edition is the critical representation of historic documents that often stand for a certain text or work. But what then is a digital scholarly edition? It has been said that digital editions are essentially different from printed editions in their content, structure and role. Yet, they share the same subject and have the same goals. Because of that, we can stick to the same general definition. The difference is not so much between editions and digital editions, but between the various forms of editions. The scholarly edition as we know it from the last several centuries is the printed edition, but the changes in our technological and media environment make us aware of the fact that there is an alternative to the print edition. The print edition is no longer the edition but becomes recognisable as a particular form. Therefore, the basic definition of the scholarly edition is valid for both varieties and we have only to discuss the difference between the printed and the digital edition. To do so, we could name the distinct contents and features of them. Or, we could describe in detail how they deal with the representation and critical treatment of their material. Digital editions already at first glance display additional, specific, characteristic aspects. Some of them can be gained by transforming printed editions into electronic texts and digital publications. Here we may talk about accessibility, searchability, usability and computability. But there are other, more essential aspects of digital editions that stem from a change in the praxis of preparation, in the methods applied and in the underlying theoretical assumptions. It can be said that digital editions follow a digital paradigm, just as printed editions have been following a paradigm that was shaped by the technical limitations and cultural practices of typography and book printing. With the mere digitisation of printed material, the implications of a truly digital paradigm cannot be realised. Digital imaging of source documents and the potential of digitally encoded text can be named as two examples of this phenomenon. As for the former: while printed editions, due to economic restrictions, usually come without facsimiles as a visual counterpart to the typographic text, digital editions usually start with visual representations, are indeed expected to provide this evidence, and where they do not, they need to justify the absence of this feature. As for the latter: while printed editions normally give exactly one version of a text, the deeply marked up textual code of the digital edition theoretically covers several views of the text and may lead to various presentations generated by specific algorithms. This fundamental difference in paradigm and its consequences for the reality of editions in our digital media landscape lead to the following important conclusion:
A digitised edition is not a digital edition.
15As long as the contents and functionalities of a typographically born and typographically envisioned edition do not really change with the conversion to digital data, we should not call these derivate editions ‘digital’. It is the conceptual framework that makes the thing–not the method of storage of the information either on paper or as bits and bytes. We can make this more productive in a more definitional manner by stating that:
A digital edition cannot be given in print without significant loss of content and functionality.
16Of course, the content of digital editions can–in theory–be printed out. And, of course, the text of digital editions could still be read on paper. However, a main characteristic of a digital edition is its representation of a potentially large number of documents in a potentially limitless number of different views, such as facsimile, diplomatic transcription and reading versions. All are generated from the same electronic code according to certain, sometimes even user controlled, modulations. The same holds true for functionality: there is no simple search, no advanced search, no real interactivity, no control over behaviour and appearance, and no source code download in printed editions. There are fewer browsing paths, no real hyperlinks, and no integrated technical tools. That is why digitisation may change the accessibility of a printed edition and may add at least some basic functionalities such as searching–but digitisation does not make a printed edition a digital edition. There is still the difference in the general framework of the whole task. For the moment, this difference may be described rather vaguely as such:
Scholarly digital editions are scholarly editions that are guided by a digital paradigm in their theory, method and practice.
17We will have to see how this paradigm can be further described and concretised.
Aspects of the digital paradigm in editing
18Markup languages, digital media and the web have been with us for some decades now. The changes in technology and media and their repercussions on our understanding of publication, text and authorship have been under discussion from the earliest days on. We already have a large corpus of theoretical literature and well established hypotheses on the new–or not so new anymore–media environment.10 As regards scholarly editions, I have tried to give a rough conspectus of these issues elsewhere.11 Only some characteristics of our digital data and media that are important for the methodology and practice shall be mentioned here very briefly. Multimedia has been among the buzzwords of the early years but still denotes the changing relationship between text and visual forms of representing documents. In the world of printed books, it has always been easier to give a transcription than a facsimile, and accordingly text was seen as the primary form of representation, with images of documents as mere illustrations. Nowadays, even if only for the practical process of editing, projects start with digital facsimiles and subsequently create transcriptions and edited versions of the text. As for the publication, the present day user tends to expect the visual evidence as a matter of course and might be vexed by its absence. Hypertext is another buzzword from the dawn of electronic textuality. With the World Wide Web and its underlying technologies, the complex and advanced theory of hypertextuality has been reduced to the practice of simple links. However, even these hyperlinks are very momentous and mark an important difference between printed and digital texts. While the former always included rather implicit links and references, the hyperlinks of the latter restructure the contents of editions, open up new and manifold paths of reception and blur the boundaries between an edition and its contexts. The pervasive linkage between different contents and parts promote a modularised structure and a moduleoriented vision of scholarly editions. Instead of concentrating on one authoritative reading as the primary goal and content, digital editions connect various forms of representation with editorial knowledge and contextual material. This is brought to the public in the process of a fluid publication in a double sense. What we see on the screen is often generated in real time from the current state of data, representing the current state of the editorial knowledge in a project. This is one aspect of fluidity. The other is the loss of a distinct moment of publication. Release early–release often! The edition loses its recognisability as an authoritative, final statement. Instead, it becomes a permanent but potentially always changing documentation of an ongoing examination and processing of the objects in question. In this way, the edition as a publication is a process rather than a product. It grows incrementally not only before its final release, but also during its availability to the public. The edition as data driven fluid publication is, at least in principle, always open to change and amendments. Thus, the edition is seen as an open enterprise. In theory, it never closes down and never reaches a final state. There is always something left to do. The edition invites the team of editors and collaborators to add more material and more knowledge. As regards the people involved and the roles they play, the road leads from the single omniscient editor of the printed edition to the team of specialists with differentiated roles in the conceptually and technically complex digital edition. It leads to a social edition, where input comes from within the team and from outside. Contributions are made by external institutions such as libraries and archives but some editions also try to attract and activate the communities of the scholarly or even wider interested public. In the end, the practice of crowdsourcing makes everybody a potential editor or at least a contributor to a fundamentally collaborative endeavour.
19Amplification and change of functionalities is one of the most obvious aspects in comparing traditional to digital editions. The book is a perfect device for the passive consumption of a limited amount of one-dimensional static information. Digital media, with its complex, multimedia, networked content, is in principle interactive and adaptive. It asks for more sophisticated browse and search functions to access all the material and information of an edition. A printed edition can be read. A digital edition is more like a workplace or a laboratory where the user is invited to work with the texts and documents more actively. Accordingly, in recent years we have even seen the integration of new features and tools into the edition, allowing for customisation, personalisation, manipulation and contribution. In the idea of virtual research environments, the border between primary material, its usage for interpretation and analysis, and the publication of findings is finally obliterated.
20At the heart of the edition there is still the text. But what is that text which the edition presents? In contrast to the one-text paradigm of the print edition, the digital edition shows a strong tendency towards multiple texts. As has been said, the digital facsimile, which is a representation of the text, is already a common starting point nowadays. But even the text as a linguistic entity represented by transcription is manifold. Often, editions offer a diplomatic version and a critically treated constituted edited or reading text. Sometimes texts are additionally given in translation and semantic information is pulled out and organised in a database or presented by indices. From a systematic point of view, it can be said that the representation of text is locatable on a scale of possible treatments and steps of processing of a document and its transcription. This spectrum reaches from positions that are close to the document (like image) to positions that are close to the user (like reading text), since they apply and add ever more interpretative and processing steps to the text.
21Behind the presentation of text as taking different positions in a range of all possible renditions, there lies not only the idea of incremental informational and critical digestion. Varying forms of text are not just teleologically moving toward one final goal. Rather, this conveys and embodies a pluralistic notion of text, where different information channels of documents and texts are perceived and can be represented on the level of encoded data. These textual dimensions are the subject of a pluralistic theory of text and include, to give just a few examples, the visual, the material, the scriptographic/typographic, the linguistic, the work-related and the semantic channel of information.12 It is clear that different forms of textual presentation in scholarly editions address these notions of what text actually is in different but complementary ways. As regards the digital paradigm, the expansion of the textual representation comes with the inversion of the role of the critically edited text. Within the typographic paradigm, the edited text is by far the most important feature, the core and the exclusive centre of the edition. All other forms of evidence, such as illustrative images, bibliographic information, details of script and typesetting, variant readings or semantic interpretations, are just substrata to or fortifications of it. Within the digital paradigm, the process is reversed: the editor does not write the edited text. Rather, it is developed gradually from the material documents, from visual evidence through the transcription and through the application of critical, historical, stylistic and philological knowledge. In the digital edition there is little reason to hide these other layers of textual representation from the user. But as one effect of this change in methodology, the edited text is relativised and the multiple text is facilitated.
22From a technical point of view, basic concepts of electronic texts, descriptive markup and current publishing architectures have led to what is called the single source principle. All knowledge about a text is united in a single information resource from which the publication and all textual forms within are generated algorithmically. From a conceptual point of view, these developments in the creation of digital scholarly resources can be called ‘transmedialisation’13–because today information resources are being created without primarily thinking of them in terms of publication. We are less looking forward to the layout and functionality of the presentation, but start with the decoding and encoding of what is actually there. We create information resources that are guided by abstract models and abstract descriptions of the objects at hand. The dogma of our current markup strategies is the separation or rather translation from form to content. Thus, we do not just transform our textual witnesses from one (material) media and form into another (digital) media and form. Rather, we try to encode structures and meaning of documents and texts beyond their mediality. And from this data we may or we may not create, and from time to time recreate, arbitrary forms of presentation in one media or another. If asked what is really the gist of the matter in our still ongoing change from analogue to digital media–what ‘the real revolution’ is–my answer, at least, would be transmedialisation. The shift from media orientation to data orientation with its focus on abstraction, modelling and multipurpose representations can be shown particularly clearly for the field of scholarly editions. Here we see a transition from the edition as a media product to the edition as a modelled information resource that can be presented in media but is about the abstract representation of knowledge in the first place.
23This has consequences when it comes to the desirable transfer of editorial knowledge from the past. When printed editions are digitised, they are transformed into electronic text and digital code. As digital (re-) publications they become more easily and widely accessible, searchable and reusable. Yet, what does not change is their paradigm. The edited text does not get closer to the documents, there is still no visual evidence, no making explicit of textual structures or semantic information, limited potential for multiple views on the text. This is why a digitised edition is not a digital edition.
24Truly digital editions show some or most of the above mentioned characteristics of the digital paradigm. However, while the definition given so far helps to identify scholarly digital editions and to distinguish them from other things, there are still some open ends to discuss and some possible problems in finding the exact borderline.
25All answers seem to lead to new questions. Every definition needs words to explain its subjects, and these words in turn need to be discussed and specified. For a start, five aspects are rather arbitrarily taken up, shedding some light on open questions and areas where further thinking may be needed.
26SDE vs. DSE. Labels are important when we try to come to a common understanding of the subjects we are talking about. Some talk about Scholarly Digital Editions (SDE), others Digital Scholarly Editions (DSE). Should we use two different labels, describe two different notions, identify two different concepts and thereby construct two different things here? Maybe there are indeed two different paths in the development and creation of critical editions. A Scholarly Digital Edition would emphasise that there is the phenomenon of digital publication and now is the time to care for its scholarly quality. This would mean to add the critical dimension to otherwise potentially uncritical publications. On the other hand, the Digital Scholarly Edition would refer to the tradition and methodologies of the scholarly edition and reflect its transformation into the digital realm. Since both should lead to the same result, they do not necessarily denote different things and, beyond the detailed discussion about approaches and perceptions, both may be used as synonyms. This may hold as well for another pair of labels. As has been argued earlier, the term critical can be taken as the central definiens for what we talk about. Thus, Digital Critical Edition and Critical Digital Edition work equally well and can be used as further synonyms.
27Digital Edition vs. Digital Archive.14 Words refer to concepts. Within the current changes, even concepts seem to be in motion. Editions widen their content. When they aim at including ever more documents and finally at completeness, and when the first level of representation may be just a digital facsimile with some metadata, then the edition looks more and more like an archive. In fact, some projects that started by calling themselves editions have later changed their name to archive. On the other hand, digital archives are already critical on the bibliographic level and imply the possibility to incrementally add further critical information, other forms of representation (such as transcription) and may finally even present an edited text.15 In fact, some projects that started by calling themselves archives have later changed their name to edition.16 If we take the critical engagement and the application of scholarly knowledge as the defining characteristics of an edition, then we can say that from a certain point on, an archive starts to be an edition. However, the disparate handling of the content in a project may as well lead to the observation that some parts have the character of an edition while others resemble an archive.
28Questions of quality and thresholds. A thing can be called a scholarly edition when it is based on scholarly knowledge and critical engagement. Editions have to conform to academic standards to be accepted as the basis for further academic research. An edition gives a complete representation of its subject. Both conditions, for content and quality, raise the question from what point on something is a scholarly edition. In both cases there would be something like a threshold. And that is hard to define precisely. According to the ‘release early’ principle of web projects, an edition would be presented as soon as possible to activate its potential audience and to encourage participation and feedback. But:
An edition project is not an edition.
29Over the past decades we have seen many attempts to create editions on high levels of methodology, aiming at covering large amounts of material but eventually just fizzling out and remaining as sketches, drafts and prototypes. So where can the line be drawn between the preliminary publication and the edition? The criteria here must be content and usability. As soon as the publication makes a substantial amount or percentage of the intended documents or texts available so that it can be fruitfully used in research, we may call it an edition. The question of quality is even harder to answer, particularly in times of upcoming public, social, crowdsourced editions. Obviously, a scholarly edition comes with the promise of reliability and high standards. Digital images, transcription, textual criticism, comments, annotations and contextual texts have to substantiate the claim that this is the best possible representation of the editorial subject and that the best experts have assiduously and painstakingly applied all existing knowledge in a rigorous method. In theory, the scholarly edition is always the ‘best realisation possible’. Obviously these qualities are very hard to measure objectively. There are, however, other aspects that can be checked and verified in the evaluation of scholarly editions.17 Some of them are more functional, like accessing the edition by means of browsing and searching or the provision of registers and indices. Others are more concerned with the edition as an academic venture. They regard the editorial essentials: can the edition be cited and referenced? Is it determinable bibliographically because responsible editors and place and time of creation and publication are indicated? Are the basic assumptions, the theory behind the edition, the methods and procedures of transcription and criticism of the text stated clearly and applied transparently? The most basic exigency in traditional editing–State your rules and follow them!–is as well the central law and starting point of all digital editing. As for the quality of these rules and their application, again the question of usefulness is crucial. Does the edition provide a reliable proxy for the documents? Can scholarly research be trustworthily based on the edition without the need to go back to the originals?
30What is one digital edition? Talking about editions, evaluating editions and cataloguing editions requires their identification by external boundaries and internal constituents. This was not a problem in the world of books. The edition was limited and identified by binding, covers, bibliographic description and as a stabilised product. The digital edition claims to be a process rather than a product, and is thus unstable as regards publication and changing content. Furthermore, the editor as fixed point is weakened by the larger teams of specialists, collaborators or even the public contributors in the social and crowdsourced edition. The publication, as algorithmically generated from separated underlying data, becomes arbitrary. Data, even from distributed sources may fuel various editions, differing in scope and distributed over place and time. Editorial content is transformed into modules or even more fine granular sets or particles of addressable, linkable and integratable objects. Editorial projects serve as platforms and portals featuring single works that are processed and annotated in depth and presented with particular functionalities. All of these phenomena make it hard to decide what forms a or one digital edition in the end. It seems that a solution can only come from the editors themselves. They set the framework and define the subject. If they declare something to be the edition we may follow them. As in the old world, the edition can then be defined as a bibliographical object that should clearly be identified and described.
31Publication vs. data. If the edition is arbitrarily created from abstract data and may be recreated by others in different forms at any time–is the edition still the publication or is it rather the data behind the publication? It seems that the data is the place where the editorial content is stored, where the editorial processes are recorded and the editorial knowledge is kept. The most important task for the editor is the creation of information as rich, accurate and reliable data. The creation of online publications or print spin-offs from this data may be left to other specialists such as publishing houses, web agencies or media designers. In fact, editions are produced in this manner. Nevertheless, it may still be the editor who is responsible for the edition as a publication as well and thus identified as the creator of it. This is surely true for, but may be restricted to, those cases where the editor is the head of the publishing process deciding on the selection of material, its presentation as well as features and functions for browsing, searching and using the edition. The situation will be more complicated when another publication is created by somebody else. Decisions on the arrangement of a new presentation may be called editorial as well, and perhaps we have to differentiate at least two layers of editorial activities: creating data and creating an edition as publication. And there is even a layer in between. Following the idea of edition as mere data, it would best be provided via formalised web services for harvesting or direct integration into more presentational forms with graphical user interfaces.18 Such services would be another, and perhaps the most adequate, form of publication.
32There is nothing as practical as a good theory. An attempt has been made here to propose a broad, but hopefully clear, short and simple definition. It should be interdisciplinary, embracing all scholarly approaches, schools and materials. Most of all, it should be functional: it should give us clear guidance in how to distinguish scholarly digital editions from other entities such as retrospectively digitised editions, electronic texts, textual corpora, digital facsimiles, editorial projects, digital archives, digital libraries etc. The simple word edition, especially in the English language, can mean any kind of publication. Yet, scholarly edition refers to something else that may lead to a publication but is framed by very specific activities and is guided by a particular set of theoretical assumptions and methodologies. The scholarly edition undergoes a fundamental change that is triggered by the new possibilities of digital technologies of description, encoding and publication. Nevertheless, it takes up and takes further the basic ideas of critical editing that have been with us for some centuries now. Others may emphasise that our whole concept of editing is changing so completely that it may dissolve and be replaced by other labels. In my work, however, I see the continuity in the basic goals of providing reliable, trustworthy and useful representations of our textual and documentary heritage as the basis for further research in the humanities. My own catalogue of scholarly digital editions is an attempt to supply some empirical data for the ongoing methodological debate. To do so, I have to draw a rather sharp line between scholarly editions in a quite narrow sense and other phenomena that are also related to the manifold activities in making our cultural heritage accessible.
33When it comes to integrating a certain item into the catalogue of scholarly digital editions, I apply the definition given above by simply asking four questions:
- Is there a full representation of the subject in question? This may be an edited text or at least a very accurate transcription. Sometimes, although this is rather an exception than the rule and depends on the specific characteristics of the material, a digital facsimile may suffice. Sometimes even a structured database can be a complete representation.19
- Is it critical? Have rules for the processing of the material been stated and substantiated? Have these rules been applied in the light of the relevant scholarly knowledge on the material, its genesis, its contexts and its reception? Does the edition add information to the representation making it more accessible, understandable and usable?
- Is the edition of academic quality? Have the rules been applied rigorously and in a transparent manner? Are the responsibilities stated clearly? Does the edition suffice as a substitute for the previous editions or primary documents making it unnecessary to go back to them in most cases? Does it enable further scholarly research on a reliable and trustworthy basis?
- Does the edition follow a digital paradigm? Does it make use of the possibilities of digital technology and media? Is it not printable without a major loss of content and functionality?
34Of course, this is only one possible view on editing in the digital age. It is subjective and surely based on some biographic and geographic preconditions, but it tries to respect and embrace the different disciplines, editorial schools, materials and genres and to build a bridge between the tradition and the current changes. And, at least, it has been under permanent practical application for ten years so that it now seems ripe for further discussion and development.