Many pre-print texts emerge from an authorial context that was collective rather than individual. Pre-print texts may be the result of an accretive process, where portions of the work were added or subtracted to conform to the needs of copyist or audience, and where the surviving versions can reflect the contribution of many different minds (Fisher, 2012). A truly sensitive translator of a text from the past must take into account the historical setting of the written product, what has been called the social logic of the text (Spiegel, 1990) to interpret the words for the reader and convey the meaning for those who wrote them and for their contemporary audiences (Paul, 2008).
Just so, the job of the translator has become increasingly complex as our notion of text also expands. Ideally, the process of translating pre-modern texts would mirror their progressive creation, incorporating multiple voices and layers of interpretation. With the integration of digital modes of thinking and acting into our daily lives, what some digital theorists have called “eversion” (Jones, 2014), new textual communities in some way mimic those that participated in the collective process of pre-print text creation. Digital tools have made it possible to assemble voices in a virtual community of textual reception, one in which participants need not be in the same physical space to exchange ideas in real time about the meaning of a text.
Scholars at Fordham University have embarked upon two different translation projects that are sensitive to elements of pre-print textual creation and that aim to forefront the process in ways understood by digitally-inflected communities. The projects emerge from a pragmatic vision of work-sharing, yet simultaneously co-opt our increasing comfort with virtual communities to recreate the collective context of pre-print texts. These projects come from unrelated disciplines and cultural contexts: the first brings together scholars with linguistic, literary, and art historical expertise to produce a new translation of the Codex Aubin, a manuscript written in Nahuatl created between 1576 and 1609 by indigenous scribes in Mexico City; and the second, led by the French of Outremer Legal Texts Working Group, brings together nine scholars to examine and translate three thirteenth-century legal texts written in a type of Old French from the Holy Land.
Although digital tool-builders have recognized the power of the collective translation process from the standpoint of a division of labor, most translation platforms are designed to manage work-flow, not to accommodate incertitude, discussion, or to express a plurality of interpretations (Gambier, 2014). Using the norms of crowdsourcing as a point of departure, Fordham scholars have worked with program-developer Ben Brumfield to substantially extend his open-source digital edition tool FromThePage (http://www.fromthepage.org/) to allow groups of scholars to translate selected texts in a communal fashion. For example, since the 1841 editions of the French-language legal texts were digitized by the Internet Archive, the existing FromThePage-Internet Archive integration was extended to ingest the OCR produced by the digitization process as a starting point for page transcripts (Beugnot, 1841). The translation process itself required modifying the data model to support parallel texts for each transcribed page, as well as new user interfaces for editors to toggle between transcript and facsimile while translating. Since users may be unfamiliar with Old French or Nahuatl, the presentation interface was revised to support a translation-first experience.
Translation teams working on both Fordham FromThePage projects include language specialists, historians, and content experts who bring their proficiencies to the discussion, but who will nonetheless disagree along the way. The assembled communities are closed and highly selective; our end product will be a curated conversation that reflects the specialization of each team member. Using FromThePage gives us the option to make the deliberations transparent to our end-users in a way that remains epistemologically true by mapping the process of consensus necessary for a final translation. Reaching an agreement on our final versions, for example, may be the result of one scholar arguing for a reading according to his or her own expertise, or of a collective ignorance and the need to choose the most neutral term possible; our final translations will expose this decision-making process and situate our users in the midst of communal textual production.
Achieving this transparency required a number of unanticipated changes. The first concerned the commenting mechanism: while editors transcribing a manuscript often need to discuss unclear handwriting, translators’ discussions of variant readings need to be surfaced for readers. In addition, parallel texts required a total re-write of the full-text search mechanism to handle verbatim text of the original, editorial emendations, and English translations. Finally, the nature of the sources revealed fundamental challenges in the representation of text within FromThePage; since the page division of the Assises is an artifact of 19th-century typesetting, semantic divisions had to be de-coupled from the facsimile pages for online readers and consumers of the exported TEI-XML documents. All of these enhancements were re-incorporated into the FromThePage source code, allowing other digital edition projects to benefit from the new translation and OCR correction features. The projects underscore the communally-crafted nature of our chosen texts by marking conflicting opinions, and the results will serve as a test-case for other kinds of collaborative textual projects, particularly those that contain non-standard languages or terminologies.