A few months ago, Hardeep wrote a blog post about the importance of the XML element <choice> in our manuscript encoding tag set. The main benefit is for the Blake Archive’s search function to allow users to search for regularized spellings of words that might be abbreviated or non-standardized in Blake’s manuscripts. For example, a user searching for “Tiger” would never be directed to “Tyger” without a choice tag attached to Blake’s non-standard spelling.

Last month, I began the transcription and encoding of a typographic work titled Blake’s Chaucer: The Canterbury Pilgrims. The work is a one-page printed advertisement for a [proposed] Blake engraving of Chaucerian characters, “in a correct and finished Line manner of Engraving, similar to those original Copper Plates of ALBERT DURER, LUCAS, HISBEN, ALDEGRAVE.” Of course, the Blake advertisement here is appealing to the fame of some historically relevant engravers, but the manuscript itself only refers to them in abbreviated/non-standard/anglicized forms.

Respectively, “ALBERT DURER, LUCAS, HISBEN, ALDEGRAVE” in fact refers to:

  • Albrecht Dürer
  • Lucas van Leyden (also sometimes referred to as “Lucas Hugensz” or “Lucas Jacobsz”)
  • Hans Sebald Beham
  • Heinrich Aldegrever (or sometimes “Aldegraf”)

So obviously, in order to make our encoding of this manuscript more useful to search queries, we need to take advantage of our <choice> element that allows the standard name of each historical figure to be recognized as related to the non-standard versions used in the advertisement. Someone interested in the potential influence of Albrecht Dürer on Romantic artists and engravers would certainly want to be directed to this advertisement, and <choice> allows the Archive’s search function to make that connection.

But the fun doesn’t stop there!

Earlier this semester, our Project Coordinator Laura and the good people working on Blake’s Descriptive Catalogue came up with a few additional attributes or values that would specify how uses of <choice> were working. For now, these “types” breakdown in the following way:

  • “split_word”: a single word is divided by a line or object break.
  • “abbrev”: a word, term or name has been abbreviated.
  • “alt_name” a alternative spelling of a person’s name, e.g. “Anna” for “Anne”.
  • “work_ref”: a reference to a work that can be identified through context, but is not fully named.

We can see a couple examples of these attributes in Blake’s Chaucer.

First, an excerpt from the original manuscript:

Screenshot of an excerpt from Blake's Chaucer: The Canterbury Pilgrims.




Second, a screenshot of the [in-progress] digital transcription:

Screenshot of a transcription for Blake's Chaucer: The Canterbury Pilgrims




And third, a screenshot of the XML featuring <choice>:


Screenshot of XML for Blake's Chaucer: The Canterbury Pilgrims









Of course, a few issues arise with these attribute cases. Namely, what happens when a name is both non-standard and abbreviated? The attributes for standardized uses of “Aldegrave” are especially tricky. I hope to discuss these possibilities in our Blake Archive meeting next week.