Uncle Tom’s Cabin Project (Entry 3): Experiment

This is the third (and most speculative) draft entry in a projected multi-part posting. During next week these documents will be refined. This posting on “Experiment” is accompanied by previous posting on “Composition History” and “Editorial Approach.”

I am proposing–attempting to imagine–an editing and humanities computing experiment that would offer an alternative form of evidence for my belief that some installments of Harriet Beecher Stowe’s National Era version of Uncle Tom’s Cabin are later authorial revisions of the 1852 Jewett edition text.

During her composition process, Stowe provided copy for the National Era (initially in manuscript, but later perhaps in galleys prepared by Jewett printers), for the Jewett editors (either in manuscript or possibly in marked up newspaper printers). Leaving early manuscript drafts and considering only those intended as printer’s copy, there are the following options:

If Stowe used option 1, 2, or 4, any of which are possible, this experiment has no purpose. The trouble is, how do we know? Short of a manuscript letter that says “copied out Jewett printer’s copy from Era installment” (if someone knows of one, please tell me!), we won’t know. If I can get funds, I’d love to read all of the letters. In any case, if Stowe used option 3, I assume that the response of most people will be “duh” and a yawn. Nothing is lost except years of my time, and we may gain the difference between suspicion and having hard evidence. But if Stowe used option 5, we have something really interesting on our hands, a corollary to my suspicion (based mostly on intuition) that the Era is, at least in part, a revised copy of the Jewett edition of Uncle Tom’s Cabin. Recall, however, that option 5 increases in likelihood later in the composition process. Option 3 increases in likelihood earlier in the composition process. During middle stages of composition (October to early March) we can’t lean one way or the other. Put otherwise, the composition process used in one section of the text may not match the process used in another.

To proceed with this experiment, we must distinguish between those variants that don’t matter and those that do. In brief, the only variants that do matter are those that would not otherwise matter. That is, if a variant is likely to have originated as a conscious act, we must discard it. The only variants that can bear any weight are the insignificant ones, those that originated when the compositor followed copy unconsciously (perhaps against rules of house styling or of consistency).

By the way, the spelling of the exclamation “oh” as “O” seems not to matter to most people. But publishing houses have different styles. Jewett, for example, excludes the “h” while the Era printer includes it, but not always. In the final installments of newspaper, the Jewett spelling of “O” starts to appear regularly (Kirkham noticed this). The counts (my counts) are as follows: Jewett has 276 O’s and 22 oh’s while the Era has 8 O’s and 279 Oh’s. So the question is, if we have a pattern that seems to reflect house styling and the pattern is followed 97 percent of the time, do the breaks in the pattern indicate random noise or following copy. If all examples are considered and they seem randomly distributed, I would assume random noise during the attention to manuscript copy is the reason. But if we consider all patterns and the types of pattern breaks are clustered, we may be able to infer that the setting copy was the reason for the variation.

These types of variants must be excluded from the analysis:

• Variants in wording, presumably authorial. These include both newspaper-Jewett variants and Jewett edition variants (between uncorrected and corrected state)
• Register changes in dialect, that is, from educated to lower class or vice-versa or from biblical (thee/thou) to non-biblical or vice-versa. By the way, this variant does not always follow race; rather, the Era and Jewett edition are different.
• Punctuation variants of substance (semi-substantives), including emphasis in type styling (small caps or italics), switch between exclamation, question, or statement, or alteration of quotation form that affects sense as well as styling
• Newspaper errors, more or less obvious w/o reference to Jewett edition. Includes obvious mis-spelling, omitted or incorrect punctuation, and type damage
• Jewett errors, more or less obvious w/o reference to newspaper edition. Includes obvious mis-spelling, omitted or incorrect punctuation, and type damage.
• Unclassifiable errors.

The only type of variants included are accidentals characteristic of type styling:

• space before apostrophe in contractions
• quotation indicated but signed differently (e.g., em dash vs. quote marks)
• Hyphenation of compounds
• Indifferent spelling variants, including dialect spelling w/ or w/o apostrophes

I am trying to figure out a method to do this, and I’m drafting ideas right now. More to come.

Uncle Tom’s Cabin Project (Entry 2): Editorial Approach

This is the second draft entry in a projected multi-part posting. During next week these documents will be refined. This posting on “Editorial Approach” is accompanied by a previous posting on “Composition History” and a subsequent posting on an “Experiment.”

I am proposing–attempting to imagine–an editing and humanities computing experiment that would offer an alternative form of evidence for my belief that some installments of Harriet Beecher Stowe’s National Era version of Uncle Tom’s Cabin are later authorial revisions of the 1852 Jewett edition text.

Editorial Approach
1. While some editorial approaches (such as German genetic editing) generally hold that one respects authority of individual documentary record (book has one version of text, serial has another, no mixing), other approaches (Greg-Bowers-Tanselle) hold that authorial intention is a valid historical interest and that authorial revision would tend to place greater emphasis on larger concerns (wording) and less interest on punctuation and type styling. This is not true of all periods or all authors, but I think (based on my review of the UTC manuscripts) there is an element of truth in it for Stowe.
2. Hershel Parker in Verbal Icon privileges the revision moments most nearly associated with active composition. If an author botches a text (maims it) because an editor complained, Parker thinks the study of authorial intention should reject those changes. I think he goes a little far, and a skeptic might consider this a divine afflatus theory of composition, but it has some merit. Those changes made months later may no longer reflect active engagement with the text. Revisions easily muck up ideas that may have been better presented in an unrevised text.
3. What do scholars need? A few years ago, I thought that an edition of the newspaper version of Stowe’s text was needed most. So that’s what I did in my dissertation. Now I think a study of the variants is needed most. That’s why I am proposing a fluid text edition along the principles advanced by John Bryant. While the edition would also include scrupulous documentary versions of each text, it would also include an editorially prepared text.

Uncle Tom’s Cabin Project (Entry 1): Composition History

This is the first draft entry in a projected multi-part posting. During next week these documents will be refined. This posting on “Composition History” is to be accompanied by a posting on “Editorial Approach” and another on an “Experiment.”

I am proposing–attempting to imagine–an editing and humanities computing experiment that would offer an alternative form of evidence for my belief that some installments of Harriet Beecher Stowe’s National Era version of Uncle Tom’s Cabin are later authorial revisions of the 1852 Jewett edition text.

Composition History
1. From early 1851 to September 1851, Stowe composed her manuscript of UTC primarily with serial publication in mind. Hedrick, for example, has observed the shift in method in chapters 14 and 15, which instead of dialog have “unbroken blocks of narration” (Life 222). This would make sense. The Era’s 8 May 1851 editorial notice projected that the length of the Stowe’s story would match Southworth’s Retribution (Era, 1851:74). Southworth’s tale had run for 14 installments.
2. From September through mid-January (and through remainder of process), Stowe had to compose with both forms (book and periodical) and audiences in mind.
3. From at least mid-January (and probably earlier) through early March 1852, Stowe composed primarily with book publication in mind–as amount of serial text published gradually fell behind book schedule. In the 12 February 1852 installment (chapter XXXII in Era, XXXIII in Jewett edition), textual variants increase in frequency and significance. Kirkham argues that the cotton-weighing episode is “much better described” in the newspaper (169), but he believes that the Era has the “earlier” version. (169-71). If Stowe’s composition of Jewett text had advanced beyond serial, it is possible that the newspaper version is in fact a later version than the Jewett edition text.
4. Stowe hoped to complete the manuscript of Uncle Tom’s Cabin in February and completed it on 2 March 1852, according to letters in the Massachusetts Historical Society.
5. Throughout March 1852 Stowe had the opportunity to revise book publication form for serial publication. The texts differ markedly. It’s partially a matter of deciding whether the differences are best described as “revisions” or as alternate texts for different audiences.

Error and the Textual Scholar

Early in my graduate school career, I was affected deeply by a comment from textual scholar Peter L. Shillingsburg. He asked, about some work that I was doing–and I paraphrase to the best of my memory–”If your quotation of the work absolutely had to be correct, would you cite this edition?” He referred to an edition that had been prepared on the basis of microfilm copies. The procedure in that edition was that the text was keyboarded and carefully proofed. When microfilm was not clear, original was checked.

On thinking long and hard, I decided that the answer was no. I would not trust the edition. Although the edition was prepared with typical standards, I silently swore to myself that from here on out I would always attempt nothing less–in editorial work–than accuracy of transcription with excruciating attention to detail. I would here remark that it requires considerable self-discipline and knowledge about tools. One can never, ever take an easy way out. And one has to understand how to use electronic collation software.

Therefore, editorial work requires at least two independent keyboardings. When I worked on Era UTC, I performed one keyboarding myself, I compared every keystroke to a text prepared by a keyboarding prepared for another electronic text transcription project (Accessible Archives). Every variant character between the two keyboardings was individually checked against an original printing and corrected. Then I and a partner orally proofed the corrected text. To test the accuracy of oral proofreading, I had someone insert errors in the text. During oral proofreading, we caught approximately 80 percent of the inserted errors. And thus based on the perhaps shaky assumption that the rate of catching planted errors matched errors in two keyboardings, I was able to estimate the number of errors that remained in the text. To discover a missed error in the oral proofreading was horribly disappointing–every time–and to have an estimate of the errors that remained was also horrible. I also compared text to a third keyboarding (of the Jewett text). Every time I corrected an set of errors, I always corrected them in two places, to confirm that the correction made properly. For full details of procedure, see chapter 1 of my dissertation.

What I did not do, when all was said and done, was to go back and silently proof the entire text. I was stymied by the two versions of the text on my site, and to proof the public text would require two silent proofings. As the public version of my site was only available for about three weeks before my defense, I went on vacation and put off another proofing until after dissertation was done. Being a human, the impulse to go back and re-proof the text twice (but also thinking philosophically about whether I have a responsibility to keep text in same form in which it was published), when silent proofreading is so mind-numbing and the work that is more likely to provide professional advancement is writing, has been postponed until I need to update the text.

I’ve proofread two installments since and found one error. The individual error is a case where I made a mistake in transcription, Accessible Archives made the same one, I missed it in oral proofreading, and I did not catch it when comparing the Jewett text, probably because I had another error in the same phrase, a reversed quote mark. Mistakes are not unlikely in a text of 700,000 characters. The error is this. When Eliza says

[...] give your consent—to—to”

the last “to” should be followed by an em dash. But it is annoying because one error per two installments would be a higher rate of error than I estimated in my edition.

I am currently reviewing all of the variants in between the Jewett text and the Era. Based on initial work, I begin to think I’m going to find 30 or more variants that by some reasonable criteria would count as errors in the Jewett text. So, onward.

Canonization and Reprint Culture, a note on McGill

I’ve been immersed in Meredith McGill’s formidable American Literature and the Culture of Reprinting, 1834-1853. McGill provides a compellingly readable account of antebellum reprint culture. The book’s method illuminates a disciplinary boundary and a fundamental paradox. The paradox is that the formal tools of canons (bibliographies, author logs) are a necessary precondition for modes of inquiry that unsettle the rationales that form canons. McGill’s willed blindness enables insight, as she is aware and clearly explains, but what insight can we gain from the blindness that seems present but not critically considered? To wit, that texts are not as stable as her readings suggest.

I’ll begin with McGill’s insight, which she explains clearly in her introduction: “For both rhetorical and practical reasons, I will focus in this book on the careers and writing of canonical writers. Like the reprinters themselves, I trade on the cachet of major authors, hoping to make the lineaments of reprint culture as visible as possible. But I also depend on these authors’ canonical status for the bibliographic research that undergirds this study. The undisputed importance of Dickens, Poe and Hawthorne to British and American literary history has made it possible for me to begin to recover the patterns of reprinting of their texts” (41). McGill excludes Nathaniel Parker Willis and Catharine Maria Sedgwick–though she identifies them as “more important to reprint culture” (42)–because they lack the requisite “bibliographic research” that is necessary to identify texts for study.

The discipline in which McGill positions her study, history of the book, the American branch of the tradition whose authorizing voice is Febvre’s L’apparition du livre, can engage in the most far-reaching and thoughtful assessments of reprint culture if it borrows from the canon-establishing fields of enumerative author bibliographies. What in particular must be borrowed is knowledge of reprint text forms for the authors that are the subject of McGill’s study. If no bibliography has been prepared, there’s no subject to research. Almost all bibliographies that aim to treat a topic in definitive fashion are prepared under the sign of an author’s name. Actually, there is another subject for research. One could build a bibliography of Nathaniel Parker Willis or Catherine Maria Sedgwick as reprinted authors, as a test of whether McGill’s insights drawn from “canonical” authors will pan out for the “more important” authors in reprint culture. A hard slog. One might note that books of criticism can earn tenure whereas bibliographies in preparation for the past decade litter the road.

The aspect of reprint culture that emerges in book history as seldom more than an interesting aside is textual alteration. [Note: A general principle, or "fact" in the words of John Bryant, is that texts, when converted from one form to another, are altered.] McGill routinely examines one type of textual alteration. If the title is altered, the alteration is remarked and commented on. But as for those smaller changes that demand painstaking examination, McGill shows little interest.

For example, McGill explains that Poe’s “Fall of the House of Usher” as published in Tales of the Grotesque and Arabesque (1839) was picked up and published anonymously by London’s Bentley’s Miscellany and then republished in the Boston Notion as a British import. McGill assumes that the Boston Notion editors are “unaware that the tale was Poe’s” with the justification that they had printed a “scathing review of Tales of the Grotesque and Arabesque in December of 1839″ (157). The foot note for this statement refers the reader to the Poe Log, page 282. This reference is obfuscatory. Why should we assume that the Boston Notion editors do not know the story is by Poe? Because they reviewed his collection? And they would not know because Tales of the Grotesque and Arabesque (1840) has Poe’s name on the title page? See Documenting the American South. This mystery is explained in a foot note to a different part of McGill’s book, number 26 on page 322, which begins “Although many readers undoubtedly knew Poe to the the author…” and eventually gets around to accumulated general evidence to the contrary. McGill cites three letters (from Poe Log) from readers who did not know of Poe’s authorship in particular cases. If some readers did not know of Poe’s authorship, then we make the large stretch that Boston Notion editors did not know of Poe’s authorship of a story from a book that they reviewed? Perhaps, but the preponderance of McGill’s evidence suggests the opposite.

McGill’s story of Poe’s reprinting, like the stories about Dickens and the history of copyright, are mostly informative and well told, but the Poe text is unlikely to have crossed the Atlantic twice without alterations. Maybe all of the alterations were insignificant. Had McGill written, “textual alterations were slight,” four words which could be written following many hours of painstaking labor (the number of hours depends on whether electronic texts of particular versions have been prepared, one’s facility with tools, etc.), then I’d know that the texts had been compared. But the 4-word phrase is not found because book history too often avoids detailed textual comparison. Book history has a broader appeal to cultural critics.

Despite my reservations, McGill’s work is important. Most scholars of American literature and culture would benefit from reading it. I did. As for another infelicity, McGill’s reference to the National Era as a magazine… oh well. If McGill had examined an actual copy, she would doubtless not have mistook it for a magazine. But since she appears to have relied on secondary scholarship, the mistake reveals that this portion of the study was not based on examination of the periodical as a physical object.