Considering the Scholarly Edition in the Digital Age: A White Paper of the Modern Language Association’s Committee on Scholarly Editions

Preamble and Statement of Purpose

This white paper is intended as a tool for thinking through a set of pressing questions for the MLA’s Committee on Scholarly Editions (CSE) and as a contextualized expression of our current responses to those questions. These questions at bottom amount to “What is a (digital) scholarly edition?” and “How can the CSE, through its practices and guidelines, encourage excellence in (digital) scholarly editing?” These questions are of course not new, but the committee has not yet addressed them directly and formally in the context of digital editorial practice. This white paper is an attempt to do so.

The audience for this document is threefold. First, it is intended for the present and future members of the CSE, for whom it will hopefully serve as a record of current ideas and rationales and  help in anchoring policy or explaining decisions to later committee members. Second, it is intended for scholarly editors who are interested in using the CSE’s guidelines and would like to know more about their intellectual background. And finally, it is intended for a wider audience that includes those who are interested in scholarly editing and editions and are curious about the evolution of the committee’s thinking.

The main focus of this document is definitional: we examine in turn the three major elements of the “digital scholarly edition” and explore their significance for the ways editions are read, used, and evaluated. In particular we consider a set of crucial features that we take to be fundamental to scholarly editing: transparency, accuracy, appropriateness of method, clear and responsible documentation, and the exercise of critical judgment in representing a full account of the textual situation at stake. We conclude by reflecting on the CSE’s own responsibilities in the light of these definitions. However, it’s worth posing at the outset a more fundamental question: What is the point of scholarly editions, as we currently understand them, in an era of mass data? One way to answer this question (anticipating some of the discussion further on in the document) is to note that a key trend in scholarly editing itself is toward the creation of an edition as a single perspective on a much-larger-scale text archive. The edition of Goethe’s Faust being developed at the University of Würzburg, for instance, includes a documentary archive of all Faust-related materials by Goethe and  a critical edition that draws its data from that archive. Even standing alone, these archives constitute a resource that supports large-scale textual research; if aggregated through a mechanism like HathiTrust or TAPAS, they constitute a much larger body of material that could be used for broader cultural analysis. The scholarly edition, in other words, is in some cases being rethought in a way that involves both the micro and the macro scales.

It is also important to recognize that although some kinds of large-scale cultural research can be conducted on informationally undifferentiated resources like Google Books, more nuanced research (and more powerful scholarly argumentation) requires data that are more representationally detailed. For instance, scholars studying the changing revision habits of generations of American novelists would need access to data that capture revision as an explicit informational component. The preparation of this kind of data—whether we call it scholarly editing or something else—draws on the same levels of expertise and care, and the same kind of attention to the specificities of texts, as the traditional scholarly edition, albeit applied toward new ends; this is a particularly fertile area for future collaboration among those in scholarly editing, bibliography, information studies, and the digital humanities more broadly construed.

Finally, along with the trend toward scale we are now also seeing a concomitant acknowledgment of the interdependence of micro- and macroanalysis, and we are seeing increased emphasis on approaches that enable scholars to move effectively between the two. The scholarly edition as we see it emerging here is well adapted to both kinds of work: it offers a detailed account of the data scholars need in order to make sense of a specific textual landscape, but it does so in a way that is formalized and programmatic and hence can support computational analysis at any scale. Through the use of standards like the TEI Guidelines, editions can also be studied in groups (though clearly this requires careful coordination of efforts to make the data commensurable across editions).

Issues

We proceed by first considering how to define the central terms (edition, scholarly, and digital) for purposes of this work and then considering the minimal qualities a “digital scholarly edition” ought to have, how we can enable these qualities, and what further research questions arise that might further the field.

Edition

Our definition of an edition begins with the idea that all editions are mediations of some kind: they are a medium through which we encounter some text or document and through which we can study it. In this sense an edition is a re-presentation, a representational apparatus, and as such it carries the responsibility not only to achieve that mediation but also to explain it: to make the apparatus visible and accessible to criticism.

To unpack this further: an edition is a systematic account of a text guided by a specific theory of what such an account should be (e.g., one that is concerned with the genesis of a literary work, one that is concerned with the social ecology of the text, one that is concerned with contextualizing a single unpublished manuscript document). An edition is thus also a model, in the sense that it serves as an analytic surrogate for the textual landscape it describes, one that can be manipulated and queried to yield insight into its details. Although this definition sounds as if it might apply chiefly to digital editions, in fact it is also true of print: there the manipulation in question may happen through the creation of multiple views of the same data (for instance, indexes, bibliographies, concordances). In print these are necessarily represented as distinct entities, but they constitute the same kind of approach to the data.

There are a few other aspects of a potential edition that are more controversial. If we consider the facsimile edition as a kind of limit case—one that takes to one logical conclusion the idea of putting curated textual materials before the reader within a framework that permits analysis and interpretation—then we need to consider what kinds and degrees of curation constitute editing for purposes of our definition. Among other things this thought experiment encourages us to consider the role of the edition as a model of a textual space that makes its contents tractable to analysis rather than an aggregation that minimizes its mediation of those contents. Another issue is that of comprehensiveness and whether an edition needs to represent and curate all extant texts and documents. While comprehensiveness is desirable to the extent that it puts the reader in possession of a maximum amount of relevant information, there are certainly legitimate editorial situations where a focus on a single document, or on a limited subset of available documents, may be appropriate. In addition, in the digital medium an edition may in fact be a specific view of a larger set of materials.

Scholarly Edition

Differentiated from other types of editions, a scholarly edition is one that follows scholarly method and purpose, that is undertaken with professional critical judgment and the fullest possible understanding of the relevant primary materials, and that provides clear documentary evidence of the relations and contexts of those primary materials. It is transparent and explicit in demonstrating an attention to the methods of its creation pertinent to the textual situation of its contents and evolving scholarly practice, in documenting the processes by which it was created, and in attending to the concerns of its medium or media. It is typically prepared with an audience of scholars and students in mind, although it may in fact serve a much broader audience, and it may also have pedagogical aims related to how it presents information and supports learning. The rigor of its preparation is assured through qualitative review, with attention to the application of, or critically constructive relation to, best practices; demonstrated historical knowledge and editorial method; completeness and accuracy of textual account and resultant text or texts; pertinence and utility of textual apparatus and paratext; and other factors relating to its scholarly reliability and usefulness.

A scholarly edition is clear about its commitments, and it keeps its promises. It is motivated to support further scholarship through its attention to these principles and their clear exposition, and it is understood to be part of larger scholarly enterprise, ultimately taking its place alongside and possibly in combination with similar works and allowing forms of analysis and engagement beyond those of its editorial intention, supporting further (re)mediation, (re)construction, and (re)mix in the advancement of scholarship in acts that allow, for example, the construction of other editions that may explore alternative hypotheses or challenge notions of authorial intention and editorial authority.

Digital Edition

The digital modes in which the scholarly edition of the twenty-first century is so often expressed are deeply significant, but in many cases they serve more to realize potential already inherent in our traditional understanding of the scholarly edition than to overturn that understanding. Although the theme of innovation is common in discussions of digital scholarly editing, it is important to frame that innovation within the context of the goals and overall mission of the editorial enterprise. The digital is neither inherently a site of innovation nor a necessarily useful innovation in itself. In proposing approaches to the assessment and design of effective digital scholarly editions, this white paper therefore takes the position that the use of digital methods needs to be carefully thought through, motivated, and explained and that specific digital features need to be consistent with the scholarly goals of the edition (as articulated in the edition’s statement of method) instead of serving solely as decoration.

There are specific digital modalities that seem to us to offer particular value for scholarly editions. First, when digital editions are designed so that their textual data is captured using standards like TEI, this opens up important opportunities for alternative deployments of the data: as the basis for other editions, as transcriptions that can be compared using collation tools, as data that can be contributed to a digital repository or aggregated into a text corpus that might support quite different types of analysis. Second, the addressability of digital information (through linked data) makes it possible to support features such as user annotation, commentary, citation, and the creation of additional layers of editorial information. Third, the emphasis on writeability, which is so important to modern digital interfaces, also extends to theories of the digital edition: edition interfaces can serve as environments for manipulation and exploration of the edition’s textual space and also as environments within which the user can occupy the role of a contingent editor, examining less-traveled editorial paths and their interpretive consequences. This becomes especially important as one considers the alignment of emerging social computing principles and practices with those traditionally associated with scholarly editing, impacting traditional editorial authority through an emphasis on ongoing open editorial procedure and facilitation.

From another perspective, the digital offers not only additional ways of designing and building scholarly editions but also additional contexts for their use and ways of understanding their pedagogical and cultural importance. Digital communication in general requires changed ideas about literacy, entailing new skills, abilities, and dispositions in front of the activities of reading, writing, and interpreting. Textual scholarship and the study of what N. Katherine Hayles and Jessica Pressman have called “comparative textual media” (vii) lie at the heart of these new literacies, which extend across an ever-expanding variety of textual, visual, and aural media.1

Supporting the Scholarly Edition in the Digital World: (Re)Considering CSE’s Mandate

Implied in the above, a key pragmatic issue for the CSE, as well as for scholarship more generally in the area, is how we choose to define the terms relating to the past and present of the digital scholarly edition and, indeed, how we choose to view extant and emerging digital scholarly editions. Many strategies of edition definition follow traditional models and understandings of edition-oriented typology, rooted in work such as that documented and exemplified in Greetham’s Textual Scholarship: An Introduction (in “Appendix II: Some Types of Scholarly Edition”). These strategies are also evinced in the elements of the CSE’s annotated bibliography of key works in the theory of textual editing (Van Hulle). Some definitions focus specifically on the digital, attempting to extend earlier traditions and to typologize digital scholarly editing trends of the past several decades in the context of current and future work (for example, Siemens et al.; Siemens). Other approaches and examples abound, some listed among materials mentioned in this document and others in and among those scholarly editions in digital form that have been submitted to the CSE for consideration toward the  award of the committee’s seal that signifies an edition’s excellence.

The scope of this white paper and indeed the considerations it offers are necessarily framed by the historical moment and the place of the CSE, itself founded with the goal of “improving the state of scholarly editing and . . . encouraging and identifying reliable textual work,” a mission that dovetails with larger initiatives within the profession to establish scholarly editing as an authoritative basis for scholarship (“Professional Notes” 274). The current CSE is seeking to further those larger aims at a time when our understanding of terms like scholarship and editing is under revision. This white paper thus serves as an attempt to articulate the CSE’s position in relation to that revision process. Pragmatically, we must ask ourselves what we need to know to ensure that the CSE best respond to changes in the field within the scope of its mandate. In brief: What is the frame of address to allow the CSE to best position itself, through the criteria associated with the award of its seal and its potential revision, so as to ensure CSE’s continued pertinent function in the scholarly editing community? While digital scholarly editing is an area actively engaged by scholarship at the moment, and it will likely be some years until there is relatively stable agreement in the field about this, there is still much available for us to consider now in relation to the CSE mandate.

In the first instance, to enable us to evaluate editions appropriately we will need to be able to define categories of editions in nonlimiting ways that can be embraced by the CSE and its processes as well, which should reflect the ways in which these considerations are emerging in our community. A key question for us in this regard is: How can we acknowledge the plurality and evolving nature of scholarly editions while nonetheless retaining the ability to recognize excellence and failure, at the same time that we rightly distinguish failure from improper categorization or the limitations of our CSE evaluative model? A clear answer here is that, as per our guidelines, the edition needs to include a statement of purpose that the reviewer can measure against: how appropriate were the methods? how effectively were they carried out?

A further pertinent question, however, is how well our current guidelines, guiding questions for reviewers, and other supporting resources reflect the current and anticipated needs of the future. It has been contemplated that these documents would require revision to accommodate this and other issues. The CSE recognizes that the category of “edition” is extremely broad and sees its own mandate as encouraging and cultivating standards for excellence within that domain. However, at its inception and in its early documents the CSE adopted a fairly specific definition of the kinds of editions it would cultivate and endorse: critical editions in the tradition of Fredson Bowers. This definition tended to exclude editions of other sorts: for instance, documentary editions. The more recent CSE discussions have emphasized the need to broaden the scope of the CSE’s attention to include different editorial modalities. These discussions, however, have highlighted the need for a set of standards of excellence that can generalize well across different types of editions.

To this end, we have arrived tentatively at an initial set of minimal conditions that mark an edition—in our terms—as a scholarly edition now, extant across modalities that could not possibly have been anticipated at CSE’s inception or in some cases even a decade ago:

  • it must account completely and responsibly for the textual landscape it represents;
  • it must fully describe and justify its editorial methods;
  • it should reveal the processes by which it was created and disseminated (including data, data structures and constraints, and algorithmic or dynamic processes), and it should include a record of changes and updates made to the edition over time, which otherwise tend to remain invisible in the digital environment;
  • it should reveal the judgment and scholarship, the editorial rationales and processes, on which the edition is based;
  • it should evince a rigorous standard of accuracy and consistency in applying a particular editorial approach, set of theoretical premises, or method;
  • it should demonstrate the appropriate fit between stated methodology, stated goals of the edition (e.g., reconstructing authorial intent, reconstructing the social text, etc.), and the nature of the existing textual witnesses;
  • it should contain a detailed textual introduction or editorial policy statement, as distinguished from a critical introduction, that outlines these aspects; and
  • it should include consideration of how the edition can circulate and function as a scholarly resource over time.

Further conditions that apply specifically to a digital scholarly edition include, but are not limited to, the following:

  • it must note its technological choices and be aware of their implications, ideally using technologies appropriate to the goals of the edition (see fit between methods and goals, above), in recognition of the fact that technologies and methods are interrelated in that no technical decisions are innocent of methodological implications and vice versa;
  • it should be created and presented in ways ensuring greatest chance of longevity — a challenge the address of which involves infrastructural, financial, and data representation issues (such as the use of widely accepted, open standards);
  • it should readily respond to the challenge of maintaining the scholarly ability to be referenced in view of the ways that interfaces change over time; and
  • where possible, it should attend to possibilities of sampling, reuse, and remix, supporting approaches to the formation and curation of the edition such as reconstructing and documenting instances of texts and textual change over time, like algorithmic construction and reconstruction (with possible extensibility, including external data); in doing so, it should attempt to balance considerations for intellectual property and labor with the goals of achieving open access and reusability.

Additional criteria may emerge with further discussion and consideration.

In closing, we wish to reiterate that the CSE remains open to and encourages the practice of a wide variety of editorial approaches, as these relate to both print and digital editions. As its primary mission, the CSE seeks to encourage excellence in scholarly editing, by which we mean above all:

  • transparency with respect to data and methods
  • clear articulation of motives
  • persuasive rationale for the editorial approach taken
  • thoroughness and accuracy
  • attention to issues of usability of the edition, including questions of audience and of long-term usability

These criteria should be understood as applying equally to print and digital editions. If we take printed books to be “machines of simulation,” as Jerome McGann has recently put it (93), this vantage point may help throw into relief the extent to which both types of scholarly edition share overlapping motives, processes, and outcomes, even while we bear in mind their clear and inevitable differences in conception and execution.

For the foreseeable future, at least, there will continue to be editions that exist only or primarily in print as well as those that exist only or primarily in digital form, with the choice of editorial format responding to pragmatic, theoretical, and (in the case of editions affiliated with a university or commercial press) marketing considerations. Print editions will benefit from established practices of marketing and publicity, quality control overseen in part by a press, and proven means of distribution and long-term preservation. At the same time, print editions lack the ability to incorporate seamlessly new discoveries after work has been finished, to make corrections, and to take advantage of the many other features of digital editions discussed in this white paper. Print editors would do well to think about the affordances of the digital, including data sharing. Significantly, even print editions now have a digital workflow. But there are also important editorial and interpretive aims that are much more feasible in print than in digital formats, as for instance McGann demonstrates in his discussion of J.C.C. Mays’s three-volume Collected Works of Samuel Taylor Coleridge (25–30). Building on Hans Walter Gabler’s edition of Ulysses, Mays’s Coleridge enables readers to toggle back and forth between “Reading Text” and “Variorum Text,” privileging neither through the editorial presentation or apparatus but instead encouraging readers to perceive the poems “in a permanent state of multiple vision” (115). Some materials from print editions might also be made available for nonconsumptive use by scholars in the digital humanities, in ways that would allow for text analysis and data mining, for instance. High-quality editions thus continue to be produced in both print and digital forms. The CSE will remain committed to encouraging and discerning the best practices for multiple types of editors and with multiple categories of readers in mind.

Note

  1. Revised in September 2015. The original sentence read, “Textual scholarship and what N. Katherine Hayles and Jessica Pressman have called ‘comparative media studies’ (vii) lie at the heart of this kind of new literacy.”

Works Cited

Greetham, D. C. Textual Scholarship: An Introduction. New York: Garland, 1994. Print.

Hayles, N. Katherine, and Jessica Pressman. “Introduction: Making, Critique: A Media Framework.” Comparative Textual Media. Ed. Hayles and Pressman. Minneapolis: U of Minnesota P, 2013. vii–xxxiii. Print.

Mays, J. C. C., ed. Collected Works of Samuel Taylor Coleridge. 3 vols. Princeton: Princeton UP, 2001. Print.

McGann, Jerome. A New Republic of Letters: Memory and Scholarship in the Age of Digital Reproduction. Cambridge: Harvard UP, 2014. Print.

“Professional Notes and Comment.” PMLA 95.2 (1980): 264–76. Print.

Siemens, Ray. “Disparate Structures, Electronic and Otherwise: Conceptions of Textual Organisation in the Electronic Medium, with Reference to Electronic Editions of Shakespeare and the Internet.” Early Modern Literary Studies 3.3 (1998): 6.1–29. Web. 13 Aug. 2015. <http://purl.oclc.org/emls/03-3/siemshak.html>.

Siemens, Ray, et al. “Toward Modeling the Social Edition: An Approach to Understanding the Electronic Scholarly Edition in the Context of New and Emerging Social Media.” Literary and Linguistic Computing 27.4 (2012): 445–61. Web. 12 July 2012. <http://dx.doi.org/10.1093/llc/fqs013>.

Van Hulle, Dirk. Electronic Textual Editing: Annotated Bibliography: Key Works in the Theory of Textual Editing. Text Encoding Initiative. TEI, 31 Oct. 2007. Web. 13 Aug. 2015.

 

The Committee on Scholarly Editions

June 2015, revised September 2015