Silence seeks to provide medievalists with an array of simple, cross-platform tools that facilitate the analysis of textual variation in works of medieval verse. Usually termed mouvance or variance, this variability among different manuscript witnesses to what is ostensibly the 'same' work lies at the heart of medieval poetics. However, the present lack of personal-use tools facilitating the reading of multiple text versions in tandem encourages such variation to be conceptualized as a secondary concern.
Concretely speaking, Silence seeks to remedy this situation by offering three interrelated tools that can be employed by themselves or all together in a single workflow.
[Currently implemented]: Silence offers a template for transcribing and archiving variant versions of a medieval manuscript passage of interest. The framework employs a simplified version of the guidelines set out by the Text Encoding Initiative (TEI). Currently, this file must be edited by hand. In the first public release of Silence, however, one will be able to input and archive this material without any prior knowledge of XML or TEI.
Once this is done, the transcriptions will be displayable on your computer in a variety of useful ways. However, your work will not in any way depend on this particular set of tools: it will be exportable and understandable by any of the many TEI-enabled digital humanities resources currently available.
[Currently implemented]: When fed suitably produced transcriptions, Silence can show them to readers line-by-line. From this comes the first sense in which Silence is "silent": it can reduce the visual noise of other lines and foster concentrated reading, a must in the study of mouvance among witnesses to a textual tradition. This feature can be seen functioning below in the alpha version of the program:
In its current form, Silence contains a Python 3 script to be run from the command line. The first public version, however, will offer several different reading options from within a user-friendly graphical interface.
[Planned feature]: Silence will study the structure of your transcriptions behind the scenes and attempt to draw your attention to lines that differ substantially from the base text, highlighting them for you as you read. This will be the second and most importance sense in which Silence is "silent": it will know not to speak up in cases of common vernacular orthographic variation, understanding that one copy's qui corresponds to another copy's ki and to a third copy's abbreviated q(ui). The first version will come with support for scanning French-language variation, since this falls within my own area of expertise. However, one of the long-term goals of the Silence project is that interested medievalists working within a variety of cultural contexts will eventually provide their own 'dictionaries' of variation (in Italian, in English, etc.), situating this tool within the broader context of the 'Global Turn' occurring within the discipline of Medieval Studies.
Silence takes its name from the eponymous protagonist Old French Roman de Silence, in which a young girl named Silence is disguised as a man and raised as such in order to be eligible to inherit. Throughout the story, subtle variation in the forms taken by her name track with her shifting mode of gendered presentation: named at birth with the feminine form Silentia, she is presented to others under the masculine name Silentius, and often referred to by the neutral form Silence in the course of the text.
The program draws its name from the Roman de Silence because of the way in which this text asserts the importance of textual variation's disclosure, but with an associated element of concealment: one must be wary, it seems to say, of what must remain "Silent." In the same way, Silence--the program--seeks to express to readers the presence of variation where it is sufficiently significant, while at the same time knowing to exercise a certain silence when it is prudent to do so.
Why bother using a tool like Silence, if you already have Microsoft Word or an equivalent tool?
When faced with the question of how to study multiple versions of a text side-by-side in their private work, most scholars quite naturally gravitate toward the writing tool they most often use: Microsoft Word or a similar word-processing program. They might lay out the text in a table, create a multi-column page layout, or even clutter their screens with multiple individual files containing transcriptions of manuscript texts.
This may work fine for your needs. If so, keep it up: the best tool for the job is always the one that gets it done. However, working in a word processor carries of a number of disadvantages that can become quite troublesome in the long term.
When transcribing the text of any kind of primary source document, accuracy is of the utmost importance. When writing in the kind of plain-text files through which Silence operates, the question of accuracy begins and ends with your own effort: the software program will not intervene to adjust your meaning. In more elaborate programs like Word, however, the "autocorrect" and "autocomplete" features must manually be disabled--sometimes repeatedly, and in different places--in order to prevent overcorrection. These features are conceived to assist in day-to-day communicational functionality, where one would not want to produce a mis-spelled word or make use of unusual spacing/punctuation. In transcription activities, however, any correction attempted by the software program is likely to be harmful--and worst of all, such changes may go unnoticed by the user at the time of the transcription, leaving a record that is flawed and produces confusion down the line. Parvus error in principio magnus est in fine.
Behind the scenes, Microsoft Word and its peers fill your document files with extensive formatting notes which are broadly undecipherable by human beings, and which risk varying in their fundamental structure from one software version to the next. From this practice arises a frustrating problem that many academics have encountered, especially those who have worked on a project over a long term, or who have attempted to return to past work years after it was begun: many files produced in early versions of Microsoft Word can no longer be opened by current versions of the software; other times, needed programs simply cease to exist (as in the case of WordPerfect).
By contrast, the file plain-text structures and standards around which Silence was conceived have been in place for nearly as long as computing itself. If one were to travel back in time to the late 1970s with a file produced using this methodology, it would be readable by a human being using machines available at the time (provided, of course, that one was fortunate enough to have access to a sufficiently powerful computer). The long-term hope behind the design of Silence is that the same plain-text files will continue to remain relevant and easily readable looking forward into the 2050s, since there has never yet been a circumstance in which their relevance and decipherability in the world of computing has been interrupted.
The first complete version is set to be released in 2020. Silence is an open-source utility and will be released under the GNU General Public License (GPL).
 Paul Zumthor, Essai de poétique médiévale, Paris, Seuil, 1972; Bernard Cerquiglini, Eloge de la variante : histoire critique de la philologie, Paris, Seuil, 1989.
 The Medieval Academy of America has played a leading role in setting this agenda: click here for more information.
Joseph R. Johnson
Assistant Professor of French & Francophone Studies