This post will be brief on last week’s tasks–for purpose of self-reminding of thoughts–so that I can focus on tasks ahead.
Last week turned again to business with the process of writing a book review, which I had promised to a journal editor by September 15. This grew out of someone noticing my work on Mother Whitman, and I was asked to review Gary Schmidgall’s Containing Multitudes: Walt Whitman and the British Literary Tradition. I was rather frustrated with the book, but I did not want to write a snarky review because I respect the effort that it takes to write a substantial scholarly book. Yet I want to make substantive objections clear when they interfere with the claims that the author is attempting to make or to hint at. I tried initially to write a measured draft, but it was just impossible as my true attitude kept ruining the tone. Eventually, most of last Friday, I rewrote most of near-complete draft without censoring my objections and produced a much better draft, though knowing I would revise again to be non-snarky.
The effort to remove the more pointed phrasings on Saturday and Sunday was not going very well, as I had become convinced of what I said. In exasperation, I turned to two sets of colleagues, Robert Trogdon, my department chair, and members of my hopefully-on-temporary-hiatus NE Ohio reading group (just reached out to regular members Jon Miller at Akron, Debra Rosenthal at John Carroll, Denise Kohn at Baldwin Wallace, though Robert Nowatzki and Adam Sonstegard at Cleveland State have also attended). Jon Miller and Robert Trogdon had time to provide some excellent advice, and I revised the draft Sunday evening and yesterday and forwarded it on to the Prose Studies editor. The process reminded me that when writing it’s often necessary to get something, anything done before the real work can begin. Much real writing begins after the notes are taken and draft is complete. Also, stop underestimating how much work writing is.
The Uncle Tom’s Cabin scholarly project work was not abandoned, though the review cut into my Friday, Monday, and weekend free time. I completed the following:
- Using the Lindstrand Comparator to sight-collate 180 pages of the 1853 illustrated edition of Uncle Tom’s Cabin.
- Discovering that the encoding state of my 1853 Illustrated Edition and 1879 New Edition transcriptions were not quite up to snuff, and trying to figure out how to correct if any were without causing more trouble.
- Having Google Drive scare the bejeebees out of me when I began to wonder whether it was randomly discarding old files.
- Breathing deeply, thinking slowly, making sure I understand what the issue is before I try to tear through files with REG-EX to fix it, writing notes to myself with reminders.
- Fixing problem 2. Two copies of the illustrated edition encoding were corrected against original to conform to more recent project standard, and the 1879 New Edition of 3 copies was rigorously checked to ensure that the encoding that I thought had been made done was in fact done.
- Encoding several pages of Herman Melville’s manuscript of Billy Budd on TextLab. Cackling gleefully when I return to Stowe’s more legible MS.
- Reading quite a bit of John Locke’s Essay on Human Understanding and studying German in my free time.
- Doing student-related work and domestic chores and not reading nearly enough literary and historical criticism on Stowe, 19C, race, religious history, study of novel, etc.
- Getting angry at Kent State administration for faculty contract negotiation–because I’m an elected member of the AAUP Council.
With book review and encoding done, it’s now back to assessing the encoding standards for collation. Let me describe the editions briefly:
- The encoding work began in 2006 while working on the National Era newspaper version. The copy has more errors than a book, but it’s a relatively clean draft.
- The John P. Jewett first edition (1852) was corrected twice and type deteriorated over the first several printings that produced 160,000 copies or so. Another 140,000 copies or so would dribble out, and type would deteriorate badly. Very few errors, technically, but the declining quality of plates has led to errors in transcription, which I began to understand only after I started encoding Jewett Million edition.
- In Million edition, which Stowe revised by adding a passage in chapter 20 and seems only to interest textual scholars, I again realized compositors used spaces in contractions and before punctuation. Allowed me, for example, to distinguish colon (with top dot lost) from period and semicolon (with top dot lost) from comma. Because a cheap edition, quite a few errors, but most again concern deterioration from banging on stereotype plates. Like flesh, even metal plates are heir to ills.
- Illustrated edition: no spaces in contractions, many errors but little type damage.
- 1879 New Edition: basically illustrations from a British edition by Nathaniel Cooke and the text of the 1852 Jewett first edition, including contraction spacing again, though more late-19C dialect and word style. For example, Stowe and her mid-century printers considered a contraction one word. In 1879, contractions will have italics on one word in the contraction pair, i.e., considered as two words.
These are the complexities:
- The National Era text encoding was initially intended for PC-CASE collator, but that application only runs on 32-bit Windows system. Did it for dissertation, but I need to move up to more modern textual editing system like TUSTEP or CollateX.
- I continued to use similar encoding, which has been inflected by my use of LaTeX, to transcribe Jewett first edition, Million edition, ’53 Illustrated Edition, and ’79 Illustrated Edition.
- The plan is to use Python and REGEX to transform the encoding into a form that will work with CollateX, which will require me to be extremely systematic because my transcriptions are extraordinarily detailed: special characters, verse indentation, small caps, italics, type damage, variants, contraction spacing, etc.
- The challenge is to normalize and systematize encoding such that any loss of detail (some detail must be normalized to avoid collapsing into morass of detail about differences in type spacing) is processed systematically and documented systematically. To use a simple example, if I remove all spacing in contractions, those won’t be identified as variants. 99.7 % of readers will know right away that they don’t care, and I can describe for the 0.29% of readers who pretend to care and provide archival access the source files. When that 1 in 10,000 reader actually cares enough to check my archival source files, it’ll be there. If you want to see this done right, see Peter Shillingsburg’s statement on normalization in the scholarly edition of Thackeray’s The Newcomes.
- Get good at this digital collating process, like Barbara Bordalejo in her kick-ass Variorum edition of Darwin’s Origin of the Species.
- Describe the important variants, like John Bryant does in The Fluid Text.
- As always, say what I do and and do what I say, the motto for scholarly editors.
This is it for now. I’m starting to ramble and need to go back to the hard work. Oh, by the way, I’ve restrained myself from tweeting for almost 20 days. I’m kind of proud of myself. But I broke down and started drafting clever tweets as the damn medium seems to have taken over my mind. Logging into Twitter periodically, which I still do, feels like stopping by a bar and not drinking, except that I don’t go to bars to drink and haven’t for years and have no clue whether it’s a good comparison. Back to figuring out how to normalize and regularize, and then figuring out how to automate some? most? appropriate amount? of the process with Python and CollateX.