Open Book Publishers logo Open Access logo
  • button
  • button
  • button
GO TO...
book cover

3. Modelling Digital Scholarly Editing: From Plato to Heraclitus

Elena Pierazzo

© Elena Pierazzo, CC BY 4.0

Editing is without doubt one of the oldest scholarly activities within the Humanities. David Greetham traces its origin to the decision of Peisistratus (560–527 BC) to establish an ‘official’ text of Homer. It was this ‘suspicion’ about the authenticity of a text, he stresses, a mistrust of variants, which led to the birth of textual awareness, which in turn has developed over the last millennium and a half into the many theories and practices of what we can now call textual scholarship.1

Texts come in different versions, and variation in texts is inevitable; or, as John Bryant puts it, ‘the fluid text is a fact, not a theory’.2 In antiquity, before the invention of print, variation was mostly due to the fact that the manual act of copying resulted in the insertion of involuntary errors and innovation in the text while copying.3 The invention of print has given the impression that this problem was resolved, since all copies coming from the press were supposed to be identical; this belief has been shaken by two sets of research, however. The first considers the actual process of printing and all the possible ‘mishaps’ that characterise such a process (correcting errors, introducing errors, recomposing forms, substituting gatherings etc.).4 The second concentrates on the text itself and the editorial activity within the publishing houses, where manuscripts provided by authors were corrected, standardised and manipulated for publication.5 Another line of research concentrates on the analysis of surviving authorial drafts; for this approach the object of study is the process of writing and of authoring. The ‘final’ text is therefore considered as just one stage along the textual journey, as it is ‘final’ only until the author decides to alter it or it becomes ‘final’ with the death of the author.6 Finally, born-digital texts are versioned by definition, since a new version is created every time the ‘Save’ command is (virtually) pressed. Yet, in spite of the multiformity of texts, the editorial and, perhaps more importantly, publishing framework is in most cases the reductio ad unum, namely the conflation of variation and version into a single, authoritative text. The example mentioned of Peisistratus aiming to establish the ‘official’ version of Homer is typical: ever since antiquity the purpose of editorial work has been the production of the one, true, ‘official’ version of the text from the many that are available.

The obligation to provide a reliable text to the reader has been at once embraced and required by publishers. In fifteenth-century Italy, the first publishers consorted with leading intellectuals in order to publish the most correct text. Publishing technology in fact enabled the simultaneous production of hundreds of copies of the same text, and therefore the importance of printing a good text became paramount. In the early stages of print, the time and resources that were saved in mechanically producing many more-or-less identical copies of the same text (as opposed to hand copying them) made it sustainable to invest in the quality of texts, giving an impulse to textual scholarship, which became a commodity with a commercial value. On the other hand, one could also say that the philological skills championed by Humanists such as Lorenzo Valla or Agnolo Poliziano had found the technology able to showcase such skills and knowledge. This convergence of scholarship and technological innovation had a huge impact on the culture of the early modern period and became the vehicle for the diffusion of new religious ideas developed alongside Biblical philology.7 The publishing industry has for centuries used philological arguments to promote their products with labels such as ‘newly corrected’, ‘accurately checked against the oldest manuscripts’, ‘improved’ and ‘purged’ used as advertising, establishing a strong and long-lasting partnership with scholars. Some of these scholars became resident editors, with famous collaborations such as the one between Pietro Bembo and Aldus Manutius in sixteenth-century Venice, but also with the work of much less glamorous people (such the ones superbly described by Anthony Grafton),8 the role of which can be located in between publishers and authors. The early modern period saw a significant increase of literacy, which, in a virtuous circle, at once made print commercially viable and was fuelled and augmented by print. The new religious climate also called for a centrally controlled and established text as it was crucial for religious reformers that their followers were handed the same version of the Bible or of the prayer book. In Protestant countries the push towards direct access to the reading of the Bible also gave a strong impulse to literacy. The requirement to teach a growing number of people how to read also favoured the introduction of what Christian Vandendorpe calls the ‘standards of readability’, which include the simplification and standardisation of page layout, the unification of spelling and punctuation and the regularisation of syntax.9 Textual variation gradually became unacceptable both theoretically and culturally; and even if the unification of texts was only attained in principle—since, inevitably, textual transmission determines variation—the provision and the delivery of the one authoritative and authentic text became a goal actively pursued by publishers and scholars for different but convergent reasons.

This culture, which aimed at the establishment of the one text, contrasts with the much more varied medieval practices of textual transmission. While for some religious and legal texts there was a need to be identical to their antecedent (a necessity that still made allowance for correction and contamination), for vernacular literature it was custom to ‘acclimatise’ the text to the linguistic environment where the copied text was to be read. For example, with highly elaborative traditions such as epic or sermons alteration and therefore the creation of new versions was considered normal practice, at least for some types of texts. The textual variety of medieval texts has been seen (and in some cases is still seen) by many scholars as a problem to fix in the quest for the original text. This point of view has started to shows some cracks, starting in the beginning of the twentieth century with Joseph Bédier’s famous criticism of the Lachmannian method, and then with even more strength by the mid-1980s, with the concurrent works of Bernard Cerquiglini, D. F. McKenzie and Jerome McGann.10 In the case of Cerquiglini, in his seminal work Éloge de la variante (1989) translated into English ten years later as In Praise of the Variant, attention was drawn to variation as a testimony to the cultural environment that produced a specific copy.11 In his vision, manuscripts are no longer simply witnesses to works but witnesses to culture, and ought therefore to be studied in their own right. The work of Cerquiglini was deeply influential and lay at the base of subsequent theoretical elaborations such as ‘New’ or ‘Material’ Philology,12 with profound influences also on Genetic Criticism. In the same span of years, with a focus on modern and printed texts, McKenzie and McGann started to investigate the various agencies and social constraints surrounding textual production. Both reached the conclusion, seemingly independently, that texts cannot be seen as the intellectual product of one agent (the author), but as the often-unstable result of the dialectic interaction among several entities, with each instance of a text to be considered as a carrier of meanings of its own.

The new theories of text based on the recognition of textual variation soon found a natural medium where variation can be presented to the readers: the digital environment. Print technology (and the infrastructure around it) has developed a way of dealing with textual variation that has been considered by many as deeply unsatisfactory: the critical apparatus. The contributions of Cynthia Damon and Marina Buzzoni in this volume show that considering the critical apparatus as a way of recording textual variation is highly reductive, since it encompasses sophisticated and specialised scholarship. However, the criticism of the apparatus still stands: its highly condensed and abbreviated formalism, elaborated for a technology where space is limited, constitutes a cultural threshold only accessible by people with the highest level of education in the very specific field that has produced such a product. The indecipherability (for most) of the critical apparatus has made it easy for readers to ignore them and so for publishers to omit them in so-called ‘reading editions’ or ‘editions for the general public’; the same omission is noticeable for many digital libraries,13 the consequences of which will be discussed further below. To be fair, there are print editions that try to account for textual variation in different ways with different grades of success. John Lavagnino for instance, presents the case of the variorum edition of S. T. Coleridge’s poems edited by J. C. C. Mays, but then he agrees that such an edition is not easy to use, as it takes ‘six volumes for what a conventional presentation could easily fit into one’.14 The question here is whether editions such as this one are difficult to use because we are not accustomed to using them, or because the medium is not suitable.

The one-text culture is then the result of several convergent social, cultural and scholarly forces: the necessity of trusting texts for worship and legality, the necessity of simplifying the access to texts for people with different levels of literacy and education, the commercial viability of the delivery of texts as printed books, the scholarly engagement with textual transmission and modern readership. In particular, it is worth focusing attention on scholarly critical editions as cultural products of all the aforementioned forces. Critical editions in particular are very sophisticated forms of publication. Their purpose is to provide a reliable and citable text which is often comprised of the combinations of readings attested by different witnesses. Critical editions are often difficult and time-consuming to produce and, in the case of very complex textual traditions, can become lifelong projects, representing the culmination of the intellectual career of a scholar. As stated above, texts come in versions, and in most cases such versions are the consequence of, to use Peter Shillingsburg’s words, ‘infelicities in transmission’.15 If we consider text transmission as a form of communication, the classic Shannon-Weaver theory can be helpful in explaining this particular aspect. According to this model, a communication act is represented by the transmission of a message produced by a source from a sender to a receiver via a specific channel until it reaches its destination, using a shared code. Such communication is never perfect, however, as it can be affected by ‘noise’ on the chosen channel and by the fact that the code of the sender and the receiver can never fully coincide. So every time a text is transmitted, whether by copying it by hand or by typesetting it, something about the text changes, either because the sender (scribe, printer) more or less knowingly alters it because the medium imposes certain changes (format, types, material, commercial constraints etc.), or because the receiver (the reader, who in turn can become the next scribe of the transmission chain) misunderstands it.16 The detection and correction of these infelicities in transmission are the objects of classical textual criticism theory. Critical editors try to make sense of textual transmission and consequent variation, and critical editions are the vehicles chosen to present their understanding to the readers. Such vehicles have been elaborated and refined along several centuries and are based on a theory of the text that aspires to the reconstruction of the author’s intention, and such intention is understood to be only one. It should however be possible in theory to separate the critical activity of the editors who investigate the phases of textual transmission from the format in which such investigations are delivered. Nevertheless it is also true that scholarship, and in particular textual scholarship and critical editing, has been deeply shaped by its own delivery format, becoming in practice almost inseparable from it.

The implications of the one-text culture are widespread and profoundly shape our understanding of and expectations about the nature of texts and the way they should be presented. The provision of ‘clean’ reading texts, where almost no traces of the underlying editorial work are visible, have contributed to giving readers a false impression that stability, a ‘trueness’ of the texts, is an achievable goal, and that texts exist in a sort of pure, Neo-Platonic state, which should not be stained by editorial marks or doubts. It is reputed that Michelangelo was able to ‘see’ the finished sculptures that were hidden inside the blocks of marble, and that he thought his role as a sculptor was to free the perfect shape that was at the heart of the block, liberating it from the inert material. This poetic image, deeply influenced by the Neo-Platonism that dominated a great part of the intellectual life of the Italian Renaissance, is perfectly exemplified by some of the most extraordinary works he sculpted, namely the series of six sculptures that go under the title of I Prigioni [The Prisoners], four of which are preserved at the Galleries of Academia in Florence, while two are in Paris, at the Louvre. All these statues represent ‘slaves’ in the act of freeing themselves from the rock that imprisons them. These sculptures can be used as a metaphor for the editorial work that aspires to free the text from the debris of transmission. The underlying concepts of a text freed from impurity can even lead to the belief that, once that the editorial work has been done, it may become so authoritative as to become unquestionable. Edward Vanhoutte cites the case of Fredson Bowers, who, in a lecture delivered in 1958, objected to the suggestion that an editor should give an account of his own workings in order to allow a reader to reproduce and verify the editorial work, since ‘bibliographical research is an essential part of the scholarly editor’s tasks and thus completed at the moment of the publication of the edition’.17 Bowers then concluded that: ‘It is an anomaly for an editor proposing to establish a text to expect wiser heads to carry forward and then to apply the basic bibliographical examination of the document on which the very details of the establishment should have rested. “Every reader his own bibliographer” is an absurdity’.18 The case of Bowers can perhaps be considered an extreme, rather than the norm. In the Italian tradition, for instance, the belief that an editor can attain such a self-assured confidence in his or her work is almost inconceivable. Generations of scholars have been formed in the belief that any edition is only a working hypothesis and that the original can only ever be approximated but never attained.19 The same awareness is present in contemporary American textual theory; Peter Shillingsburg declares: ‘No edition was ever or will ever represent a work adequately. Full stop. The positive. The hopeful. The perfection. The adequacy. The triumph of scholarship. They will not occur’.20 However, even if mitigated by theoretical and methodological concerns about the effectiveness of the editorial work, the duty of the editor seems to be the production of one authoritative text (or at least aspiring to). Once the apparatus and the critical analysis of the tradition are stripped out and the naked text is proposed as a reading text, this text inevitably presents itself as The Text. This is also lamented by John Bryant: ‘the smoothness of the reading text, a hallmark of critical editing, in effect denies us an immediate awareness of the actual roughness of the textual record, and textuality itself’.21

The digital environment knows none or little of the space limitation of print. The absence of this limitation has made possible the provision of texts in versions and texts as versions in a way that is much simpler, more intuitive and dynamic than corresponding print attempts. The Samuel Beckett Digital Manuscript Project22 demonstrates this principle, as it includes all the documents that compose the dossier génétique of each of the works included, providing several visualisations for each of them and tools for exploring their differences. Another example is offered by the digital edition of De trein der traagheid, edited by Xavier Roelens, Edward Vanhoutte and Ron Van der Branden,23 where each of the nineteen witnesses that compose the tradition of this work is available on its own or in comparison with others; in addition a reading edition is also provided. Some of these digital endeavours have been labelled ‘archives’ instead of editions since they present all the witnesses of a work with or without the support of digital facsimiles. This is the case, for instance, with the Rossetti Archive or the Whitman Archive.24 The choice of labelling them ‘archives’ has generated a large amount of discussion on their scholarly nature (are they ‘critical’ enough?) and their alleged lack of textual focus.25

These editions have been accompanied by fierce discussions, with scholars arguing in particular that offering readers too many choices will only cause disorientation rather than engendering in them an appreciation of the complexity of textual variety. As stated by Robinson ‘it appears that rather few readers (indeed, rather often, only editors) actually want to see all the images, all the transcripts, all the collations’, and while ‘printed editions acted as filters’,26 the digital medium provides readers with an overload of unregulated data that struggles to become information, since ‘value […] is added through [a] chain of literary agents, specialist readers, editors and publishers’.27 Others have reminded editors of their responsibilities as providers of critical texts, rather than simply of witnesses.28 Yet it is likely that some of these critiques are connected with the uneasiness of a new textual model that is endemic in digital texts, a model based on variation and plurality of manifestations and representations.29

Digital texts can be defined as inherently variable; in fact the variance of digital texts occurs in several contexts and can assume several shapes. The first and most basic form of variation arises from the fact that each digital reader (or user) will access the text from different devices, the shape and dimension of which cannot be controlled by the editor and can vary dramatically. The experience of reading a text on the screen of a mobile phone can be radically different from that of a tablet or a laptop; the same applies to texts read from a web browser or downloaded within an eReader application or even printed out on A4 paper. All of these manifestations present the same text, yet the reading experience (and ultimately the message) can vary greatly. This is even more the case for digital scholarly editions that are offered from within a web-based user interface and often rely on specific software, the availability of which may not be possible in different environments such as the one offered by mobile devices. The second form of variation stems from the fact that many digital texts actually change. Since the digital medium allows for easy modifications, even after their first publication, many people take this opportunity and edit the texts as they go along. This ‘work-in-progress’ type of publication has become quite common if not the norm for digital scholarly editions. Many editions are in fact published quite early and in their elaboration stage in order to collect feedback and maintain the interest of the readers throughout the lifespan of the project.30 This new mode of publication offers a strident contrast to the traditional print-based publication system, which sees long delays and frantic checking before the texts are actually printed, as once one has ‘pressed the print button’ nothing can be done to fix errors that inevitably escape the most thorough of controls. This change is bound to have profound consequences for the production and consumption of scholarship.

The third form of variation of digital texts is deliberately offered as a feature of many digital editions: the possibility of displaying the same text in different ways. In fact, by the application of different sets of rules contained in the so-called stylesheets, if the text has been produced by the means of text encoding, it is possible to visualise it in any number of different formats, for instance as a critical, diplomatic, variorum or as a reading text. These visualisations can be generated dynamically on-demand by the users and form the staple of what has been defined as a paradigmatic edition,31 where paradigmatic variation lies at the heart of the theoretical set up of both the edition and its publication.

The digital medium with its inherent variability presents itself as the ideal environment to deal with text variation in ways that go beyond the possibilities offered by the print medium. Since space is not an issue, and hypertextuality simplifies navigation from one version to the next, editors have embraced the new medium in order to explore different types of scholarship able to take advantage of such variability. Coping with and exploiting this digital variability requires a rethinking of most of the heuristics of textual scholarship, as well as its editorial products. This necessity is clearly present in Michael Sperberg-McQueen’s reflection:

Editors may justly feel that electronic editions have translated them from a stable environment with difficult but well-known problems into a river of Heraclitean flux, in which everything is changing from moment to moment, and the editor and edition are expected to adapt actively to those changes from moment to moment, without being able to rely on many of the principles which used to be stable guides to editorial thinking.32

Editors and editions then have to learn how to swim in this sea of mutability, which requires:

endowing an edition not only with a store of factual knowledge concerning the work presented, but also with the capability of dealing gracefully with the mutability of the electronic medium, by exploiting the possibilities for reader-controlled changes to the edition’s presentation and by adapting successfully to rapid changes in the hardware and software environment.33

The change from a Neo-Platonic view of the text—which only needs to be freed from the errors of the transmission to present itself in its pristine status—to a Heralictean view—according to which texts are mutable by nature—has several consequences in terms of the purpose of editing, the role of the editor, the role of the reader, the workflow of editing and the types of products that we aim to publish. First of all, the purpose of editing needs to change from the provision of a stable and quotable text to the provision of an accountable reconstruction of some of the various states of the text—some, since an exhaustive account is impossible by definition in a Heraclitean framework. This does not mean that the editors do not have to provide a reading version of the text, but that this should not be the main or only focus of editing. Cesare Segre has described scholarly editions as diasystems: the system of the text and the one of the transmission of the text, or the real image of the text (the one of the author) and the historical perceived image of it.34 Traditionally, scholarly editions that present the system of the text have become predominant: the font, the layout, the distribution of text and variants on the page clearly denote a hierarchical dependency between the text and the account of the text’s transmission. A theoretical framework that recognises the intrinsic mutability of texts needs to reverse this hierarchy, however, so that the representation of the historical perception of the text occupies the centre stage of the editorial discourse.35

This change of focus of the editorial endeavour is at once fuelled and determined by the change of medium, in the sense that while a digital environment has made it possible to exploit the variability of texts, it has also invited scholars to consider doing so. Richard J. Finneran notes that the advent of new technologies ‘coincided with a fundamental shift in textual theory, away from the notion of a single-text “definitive edition”’ remarking that while ‘a traditional print edition is able to accommodate this new thinking in textual theory either awkwardly or not at all, digital technology is its necessary and inevitable realization’.36 On a similar line are Marilyn Deegan and Kathryn Sutherland, with the caveat that they describe the renewed interest in more complex considerations of textual transmission, consumption and dissemination as not being ‘computer-dependent’ but ‘computer-convergent’.37

What are the consequences for the consumption of texts? What is the impact of a dynamic, Heraclitean view of texts on their readers and users? From the perspective of the readers, this approach may represent a striking novelty which may or may not be received as a good thing. For readers who are educated by the traditional format of text delivery (the one-text edition), a digital resource that presents many versions and that showcases the transmission over the reading text may result in an unwelcomed approach; others may instead find the new format more engaging as it affords other types of reading beside the linear top-to-bottom one. Reading is an activity that takes various shapes, from the linear reading of novels and essays to the fragmentary reading of newspapers, to the information-seeking reading of dictionaries, to the intensive re-reading of objects of study. Governed by economic constraints, the publication of texts has aimed at satisfying many if not all types of reading with a single product, but as the same constraints do not apply for the digital environment, it is conceivable to produce editions that aim at only one type of reading (and readers) or editions that provide different outputs for different readers, where textual variation is not hidden away in the fear that this may ruin the aesthetic pleasure, but presented in an accessible and engaging way. Textual variation does not need to be presented as a hyper-specialised and cryptic critical apparatus, nor as a series of texts put side by side; texts and textual scholarship can be made accessible and interesting. For example, an innovative publishing house, Touch Press, has since 2010 published several apps for iPads in the field of science and literature outreach which allows for textual variation. Beside very successful apps about the periodic table of elements and the galaxy, they have produced an edition of The Waste Land by T. S. Elliot and of the Sonnets of Shakespeare.38 These apps present the texts in different ways, as video, audio, written words and any combinations of the above; they also include multimedia commentaries, facsimiles of first editions and, in the case of The Waste Land, of the typeset text as corrected by the author. It is probably not a coincidence that such an approach is pursued by a publishing house that specialises in science outreach. In the sciences there is an established tradition of encouraging the general public to engage with scientific discovery, resulting in a rapid growth in the science communication industry as demonstrated by a number of teaching programmes that have recently emerged.39 This interest in the sciences is usually created by publications written by high profile scientists aimed at the lay public40 and various initiatives that encourage a large section of the interested public into understanding science. We could also consider here science and natural history museums, for instance, where, especially in English-speaking countries, it is customary to offer science-oriented entertainment and learning activities. Such activities are also present in certain areas of the humanities, such as art and art history, as well as in history where books aimed at the non-specialist public are quite common. Other than reading editions and editions with commentaries for students, not many of these activities have involved textual scholars and scholarship, especially since in these cases the focus is usually more on understanding and interpreting the text itself rather than on a variation or transmission. The digital environment in its various embodiments may well be a more flexible space in which to seek public engagement with editorial endeavours, if indeed editors are willing to do it.

The effect of moving away from the one-text paradigm may have consequences far beyond the public of readers. Corpus linguistics, computational linguistics, stylometry, authorship attribution, data mining and many other scholarly disciplines and methodologies that approach texts from a quantitative point of view are based on the assumption that we can query the one text, and that not only does the one text exist, but also that we know what it is. For quantitative approaches the fact that texts only exist as versions is irrelevant, since the large amount of data used for such approaches normally offsets small differences and minutiae such as those that one could consider informing different text versions. For most of these approaches the one text is all that is needed, and more than one will actually only produce noise and falsify the results. This, of course, is a simplification, since we are seeing more complex approaches to textuality and textual variation being adopted by some scholars of corpus linguistics, in particular those working with historical corpora,41 but in spite of this crude generalisation it is not far from the truth that most data mining, distant reading and NLP methods assume texts to be stable, uncontroversial entities. What will the consequence of a paradigmatic change be on these methodologies? Probably none at first, since it is very likely that in most cases a cultural change such as the one invoked here will go unnoticed for a while at least. Even then—since these methodologies willingly ignore any textual complexity, adopting in most cases a positivistic view of the text—acknowledging text mutability would require a profound change in such methods, a change that seems unlikely to happen. Each discipline selects its own level of simplification and abstraction of its object of study; the simplification level that is conveyed by the one and plain text may look a step too far for textual scholars but may perfectly serve the needs of other research approaches. Furthermore, if editors do not emphasise the importance of textual variation it is hard to imagine that others will do it on their behalf.

One obvious problem is the increased difficulty in citing texts. To affirm that all academic culture is based on citations may be an exaggeration, yet the possibility of tracing and attributing a certain portion of text to a specific source is a fundamental requirement for any publication wanting to be defined as scholarly. This is not an exclusive problem of mutable texts, but a problem that concerns all scholarship delivered in digital form, including editions based on the one-text paradigm. As maintained by Christine L. Borgman ‘notions of fixity are problematic in a digital, distributed world’; yet, she continues, ‘it must remain possible to cite publications, data, and other sources in persistent ways, so that others can evaluate the evidence on which a scholarly work rests. Dynamic objects, however, will be increasingly common in scholarly work’.42 The task of coherently preserving fixed versions of online resources seems out of reach for the moment, with scholars and producers of digital resources relying on institutional repositories and on libraries and infrastructures,43 yet considerable investment and raising awareness may help in changing this situation. What is more likely to happen, however, is that scholars will learn how to cope with ephemeral objects, with innovation becoming increasingly more important than longevity. This change is indeed already taking place. The web has educated its users to check the ‘last updated’ date and to use this date as an indicator of quality and the fact that the resource is well looked after.44 Therefore when accessing completed resources such as the Rossetti Archive, completed in 2008,45 one cannot fail to feel a sense of mistrust of a resource which has seemingly been ‘abandoned’ for eight years, at the time of writing—which is, on the face of it, somewhat paradoxical, given the fundamental requirement of scholarship to be immutably and indefinitely preserved.

We are in a phase of cultural re-mediation, where most of our cultural artefacts are adapted to the new medium or substituted with digital equivalents, or allegedly so; texts and editions of texts are no exception: the examples of apps of The Waste Land and the Sonnets shows this, as well as the rapid diffusion of eBooks and tablets. Re-mediation is not a simple transposition since the medium shapes the messages in profound and unexpected ways.46 But while we will probably understand some of the long-term consequences only in the course of the next generations, we are also offered the opportunity to lead some of these changes by engaging at once with texts, the medium and the readers, to try to produce different and more complex representations of the text and text culture. Digital mutability may respond well to textual mutability but only if the latter is recognised and embraced, and if we make a feature of it rather than considering it a bug.

1 David Greetham, ‘A History of Textual Scholarship’, in The Cambridge Companion to Textual Scholarship, ed. by Neil Fraistat and Julia Flanders (Cambridge: Cambridge University Press, 2013), pp. 16–41 (p. 18).

2 John Bryant, The Fluid Text: A Theory of Revision and Editing for Book and Screen (Ann Arbor: University of Michigan Press, 2002), p. 1. A similar concept can be derived from Peter Shillingsburg, ‘Text as Matter, Concept, and Action’, Studies in Bibliography, 44 (1991), 31–83 (in particular pp. 47–51).

3 To these one must, of course, add authorial revisions which may not be as common as in the case of modern authors but are by no means absent. Even more important are re-elaborations of texts which are particularly common in certain genres, such as epic, sermons, commentaries and so on, on which see further below.

4 See, among others, Fredson Bowers, Principles of Bibliographical Description (Princeton: Princeton University Press, 1949), and Ronald B. McKerrow, An Introduction to Bibliography for Literary Students (Winchester: St Paul’s Bibliographies, 1994).

5 See, for instance, the studies over the so-called ‘accidentals’ produced following the seminal work of W. W. Greg (‘The Rationale of Copy-Text’, Studies in Bibliography, 3 (1950–1951), 19–36), a brief account of which can be found in David Greetham, Textual Scholarship: An Introduction (New York: Garland, 1994), pp. 333–35. More recently Kathryn Sutherland has studied the normalising role of publishers for Austen’s works; see Jane Austen’s Textual Lives: From Aeschylus to Bollywood (Oxford: Oxford University Press, 2005).

6 This is the approach pursued by critique génétique or Genetic Criticism, for which see Almuth Grésillon, Eléments de critique génétique: lire les manuscrits modernes (Paris: Presses universitaires de France, 1994).

7 See, for instance, Elizabeth Eisenstein, The Printing Press as an Agent of Change (Cambridge: Cambridge University Press, 1982), where the author argues for an extraordinary push towards fixation and standardisation produced by print, but also the impulse towards the democratisation of knowledge that partially made possible the establishment of the Reformation, where individual access to the Scriptures became fundamental.

8 Anthony Grafton, The Culture of Correction in Renaissance Europe (Panizzi Lectures) (London: British Library, 2011).

9 Christian Vandendorpe, From Papyrus to Hypertext: Toward the Universal Digital Library (Urbana and Chicago: University of Illinois Press, 2009), pp. 15–21.

10 This movement was anticipated by the work of Paul Zumthor, who theorised variance (mouvance) in medieval texts (Essai de poétique médiévale, Paris: Seuil, 1972); his influence has been rather limited outside the field of medieval studies, however.

11 Bernard Cerquiglini, Éloge de la variante: Histoire critique de la philologie (Paris: Seuil, 1989); English translation by Betsy Wing, In Praise of the Variant: A Critical History of Philology (Baltimore: The Johns Hopkins University Press, 1999).

12 The launch of ‘New Philology’ is traditionally related to the special issue of Speculum published in 1990, edited by Stephen G. Nichols, whose introductory essay ‘Philology in a Manuscript Culture’ has a programmatic role. For a summary of the issues, see M. J. Driscoll, ‘The Words on the Page: Thoughts on Philology, Old and New’, in Creating the Medieval Saga: Versions, Variability and Editorial Interpretations of Old Norse Saga Literature, ed. by Judy Quinn and Emily Lethbridge (Odense: University Press of Southern Denmark, 2010), pp. 85–102.

13 This absence is particularly noticeable for classical texts; see for instance the Perseus Project ( and the (Abridged) Thesaurus Linguae Grecae (

14 John Lavagnino, ‘Access’, Literary and Linguistic Computing, 24 (2009), 63–76 (p. 74),

15 Shillingsburg, Text as Matter, p. 51.

16 It is arguable that in the digital word the message can indeed be transmitted without alteration; the varying contexts of manifestation ― laptop, tablet, phone, desktop, web browser, downloaded file, etc., all with different settings and aspects to them ― do however still result in a certain amount of variation, and in spite of all effort, the level of ‘noise’ can never be reduced to zero.

17 Edward Vanhoutte, ‘Every Reader his own Bibliographer ― An Absurdity?’, in Text Editing, Print and the Digital World, ed. by Marilyn Deegan and Kathryn Sutherland (Farnham: Ashgate, 2009), pp. 99–110 (p. 99, n. 1).

18 Fredson Bowers, ‘Principle and Practice in the Editing of Early Dramatic Texts’, in Textual and Literary Criticisms: The Sandars Lectures in Bibliography 1957–1958 (Cambridge: Cambridge University Press, 1966), pp. 117–50 (p. 146).

19 See for instance the following quotation from Alfredo Stussi, which is also cited by Marina Buzzoni in the present publication: ‘l’edizione critica […] è un’ipotesi di lavoro e quindi il lettore deve essere messo in grado di verificarla punto per punto ed eventualmente di dissentire’ [a critical edition […] is, in fact, a working hypothesis and therefore the reader should be able to verify it point-by-point, and possibly disagree], Fondamenti di critica testuale, ed. by Alfredo Stussi (Bologna: il Mulino, 2nd ed. 2006), pp. 20–21 [my translation]. It is believed that the first formulation of such a principle is to be traced to Gianfranco Contini in Breviario di Ecdotica (Turin: Einaudi, 1986).

20 Peter Shillingsburg, From Gutenberg to Google (Cambridge: Cambridge University Press, 2006), p. 154.

21 Bryant, The Fluid Text, p. 27.

22 Samuel Beckett Digital Manuscript Project, ed. by Dirk Van Hulle et al. (Antwerp: University Press Anwerp, 2013),; see Dirk Van Hulle’s contribution in this volume.

23 Johan Daisne, De trein der traagheid, ed. by Xavier Roelens, Edward Vanhoutte and Ron Van der Branden (Gent: Centrum voor Teksteditie en Bronnenstudie, 2012),

24 The Complete Writing and Pictures of Dante Gabriel Rossetti: A Hypermedia Archive, ed. by Jerome McGann (Charlottesville: IATH, 2008),; The Walt Whitman Archive, ed. by Ed Folsom and Kenneth M. Price (University of Nebraska-Lincoln: Center for Digital Research in the Humanities, 1995–),

25 Kenneth M. Price, ‘Edition, Project, Database, Archive, Thematic Research Collection: What’s in a Name?’, Digital Humanities Quarterly, 3.3 (2009),; Jerome McGann, ‘Electronic Archives and Critical Editing’, Literature Compass, 7.2 (2010), 37–42.

26 Peter Robinson, ‘Electronic Editions for Everyone’, in Text and Genre in Reconstruction: Effects of Digitalization on Ideas, Behaviours, Products and Institutions, ed. by Willard McCarty (Cambridge: Open Book Publishers, 2010), pp. 145–63 (p. 150),

27 Marilyn Deegan and Kathryn Sutherland, Transferred Illusions: Digital Technology and the Forms of Print (Farnham: Ashgate, 2009), p. 115.

28 Alfredo Stussi, Introduzione agli studi di filologia italiana (Bologna: il Mulino, 2007), pp. 245–46; Peter Robinson, ‘Towards a Theory of Digital Editions’, Variants, 10 (2013), 105–32.

29 See Patrick Sahle’s contribution in this volume.

30 Just one of many examples is the Henry III Fine Rolls project that had a rolling publication of the edited text during the lifespan of the project: Henry III Fine Rolls Project, ed. by David Carpenter et al. (London: King’s College London, 2009–2013),

31 Elena Pierazzo, ‘Digital Documentary Editions and the Others’, Scholarly Editing, 35 (2014),

32 C. M. Sperberg-McQueen, ‘How to Teach your Edition how to Swim’, Literary and Linguistic Computing, 24 (2009), 27–52 (p. 30),

33 Ibid. [italics in the original].

34 Cesare Segre, Semiotica filologica: Testo e modelli culturali (Turin: Einaudi, 1979).

35 Cf. the passage, cited and translated by Marina Buzzoni in the present volume, from Cesare Segre, ‘La critica testuale’, in XIV Congresso Internazionale di Linguistica e Filologia romanza (Napoli, 15–20 Apr. 1974), I–V (Naples and Amsterdam: Macchiaroli-Benjamins, 1977–1981), I, pp. 493–99 (p. 497): Occorre […] capovolgere i rapporti gerarchici fra testo e apparato, dare la maggiore enfasi all’apparato e considerare il testo come una superficie neutra [...] su cui il filologo ha innestato le lezioni da lui considerate sicure, fra le tante considerate. Ma l’edizione si merita l’attributo di critica molto di più attraverso l’apparato, se discorsivamente problematico: perché esso sintetizza il diasistema della tradizione, e perché svolge un vaglio completo, anche se non sempre conclusivo, delle lezioni. [There needs to be a turnaround [...] in the hierarchical relationships between the text and the apparatus, give greater emphasis to the apparatus and consider the text as a neutral surface [...] on which the philologist has grafted the readings which he deemed certain among the many considered. However, the edition deserves the attribute of being ‘critical’ through the apparatus, if discursively problematic: because it summarises the diasystem of the tradition, and because it carries out a full assessment, even if not always conclusive, of the readings].

36 The Literary Text in the Digital Age, ed. by Richard Finneran (Ann Arbor: University of Michigan Press, 1996), p. x.

37 Deegan and Sutherland, Transferred Illusions, p. 64.

39 See, for instance, the MA programmes offered by the Imperial College London, University of Sheffield, University of Edinburgh, University of Cardiff, to name only a few examples in the UK.

40 See, for instance, Stephen Hawking’s A Brief History of Time: From the Big Bang to Black Holes (New York: Bantam, 1988), published many times since its first edition; according to current estimation (2016) the book has sold about ten million copies worldwide.

41 See, for instance, the Syntactic Reference Corpus of Medieval French ( and the TXM tool developed at École Normale Supérieure of Lyon by Serge Heiden (

42 Christine Borgman, Scholarship in the Digital Age: Information, Infrastructure, and the Internet (Cambridge, MA: MIT Press, 2007), p. 232.

43 See Elena Pierazzo, Digital Scholarly Editing (Farnham: Ashgate, 2015), chapter 8.

44 See, for instance, the Guidelines to evaluate digital resources issued by the University of Berkeley, which advises students to check if the resource is ‘stale’ since recent updates demonstrates that ‘the page author is still maintaining an interest in the page, or has abandoned it’ (Evaluating Web Pages: Techniques to Apply & Questions to Ask, UC Berkeley Library, 2012,

45 McGann, The Rossetti Archive.

46 ‘The medium is the message’, as famously said by Marshall McLuhan; the phrase first appeared in his Understanding Media: The Extensions of Man (New York: McGraw-Hill, 1964), p. 7.