Week 4: Encoding "England's first family of writers"

Keeping on the romantic trend I thought I would take a closer look at the Shelley-Godwin Archive that Kirschenbaum and Werner mention in "Digital Scholarship and Digital Studies: the State of the Discipline." I have a soft spot for the Shelley after taking a course on the young romantics while studying abroad at St. Andrews (where I collected a few older copies of his work before coming home). Nothing, however, compares to the collection of works that the Shelley-Godwin Archive (S-GA for short) makes accessible online. The collection provides access to handwritten material from what the archive labels “England’s first family of writers”: Mary Wollstonecraft, William Godwin, Mary Wollstonecraft Godwin Shelley and Percy Bysshe Shelley.

The archive is especially relevant to our class because not only do they use TEI to represent complex handwritten texts, but they also provide detailed documentation on how they implement TEI and what issues they have encountered in the process. What I found particularly helpful is that they address one of the same issues that we did in class this week, which is the issue of multiple hierarchies.  While this encoding issue isn’t addressed on the website itself, it is addressed in a conference paper given on the project in 2013 (which you can access here: http://digilab2.let.uniroma1.it/teiconf2013/program/papers/abstracts-paper#C168). In this paper the discussion of multiple hierarchies and the need for “milestone” elements is discussed – specifically how including such elements complicates the “correctness” of the code and the validation processes.

Looking specifically at how the SG-A puts TEI to use in their collection I was reminded of last week’s Sperberg-McQueen reading. S-GA is a prime example of how implementing a certain set of tags (or encoding a text in a specific way) embodies a specific interpretation of the text by privileging particular content and reader actions. For the content of the S-GA this is the ability to identify and trace the composition of the various documents in their collection, specifically the ability to label corrections, annotations, and the contributions of multiple authors.

Below, for example, is a page from Mary Shelley’s draft notebook of Frankenstein. The page facsimile on the left shows the page as it appears to the reader viewing the notebook, however, by “limiting view” to either Mary Shelley or Percy Shelley, their individual contributions to the page are highlighted (as Percy Shelley’s are in this example).

Screenshot of Draft Notebook A from the S-GA. accessed at http://shelleygodwinarchive.org/sc/oxford/frankenstein/notebook/a#/p11

I found that having access to the encoding of individual instances of annotation and multiple authors (or "hands") helps you to understand how different kinds of textual features can be represented using TEI. S-GA also helps the reader gain an understanding of their encoding choices by providing a list of the core TEI tags used in their collection. The inclusion of this tag set demonstrates the importance of these features to their interpretation of the texts in the archive, in addition to aiding curious readers (such as myself) identify how certain textual features are represented in TEI.

Another unique aspect of the project is that each page of a text offers a “transcription status” which indicates to the user whether or not an authoritative TEI encoded version of the page is available for the user to view and download. While not all pages are currently available, I think the transparency in encoding offered by the S-GA is honourable.

Overall, the S-GA provides a great deal of information about the project on its website, including everything from the origins of the project, its partners and contributors (there are many), and its future goals. An explanation of the technological infrastructure supporting and maintaining the archive is just one aspect, though an important one. 

References:

Sperberg-McQueen, C.M. "Text in the Electronic Age: Textual Study and Text Encoding, with Examples from Medieval Texts." Literary and Linguistic Computing 6, no. 1 (1991): 34-46.