A few weeks ago, I asked a bibliographer a question about my objection to a footnote in Kenneth S. Lynn’s edition of Uncle Tom’s Cabin (1962). In an article draft, I stated—as kindly as I knew how—that Lynn had confused two key bibliographical concepts: edition and printing. In response to my criticism of Lynn’s misunderstanding, the bibliographer suggested that I was being a bit pedantic. He reminded me that the distinction that bibliographers observe—between an impression or printing (copies printed as one set in a unit of time) and edition (all printings from substantially the same setting of type)—is foreign to most collectors, booksellers, and librarians.
Lynn in the Harvard edition stated that Stowe failed to have the Parker footnote removed “from the stereotype-plate of the first edition” (135 n. 3). If one observes the strict definitions of analytical bibliography, E. Bruce Kirkham proved Lynn’s footnote wrong in the late 1970s when he showed that the stereotype plates for Stowe’s two-volume Jewett edition were altered twice, once after the first printing of 5,000 copies and before the printing of the 10,000th copy and once at or around probably the eighth or tenth printing that included copies labeled 50,000th. If one defines bibliographical terms strictly, my reading is correct. But Lynn never claimed to be a bibliographer. If a nonspecialist says first “edition” while meaning first “printing” it is ungenerous of me to call out the error from a narrow disciplinary framework.
I use this anecdote to open my response to Les Harrison’s call for a dialogue, because I don’t want to again slip into pedanticism, as I probably did when I suggested that “dismissing books as physical objects and dismissing the intellectual rigor necessary to reproduce them as someone else’s field (or mere workmanship) is to narrow the scope of digital humanities.” I slip into pedanticism too easily, as perhaps Les did when he named our digital chat a “dialogue,” a word with which he may have intended to scare away general readers–probably as well, as our subject is digital humanities as a topic of research.
I recall instances in which the struggle to think clearly is fruitful, even when the form of expression may be pedantic. For example, in my dissertation, I never once called John P. Jewett’s two-volume 1852 edition the first edition of Uncle Tom’s Cabin because most installments from the National Era version preceded Jewett’s book, one installment was released simultaneously with the two-volume book, and the final newspaper installments were released after the two-volume book. Hence, in paragraph above, I say “Stowe’s two-volume Jewett edition” for what everyone thinks of as the “first” edition. But I think that the first three-quarters of the Jewett edition is a revised version of the National Era text while the last quarter of the National Era text is a revised version of the Jewett text. About a year after I finished my dissertation, I discovered that I was not the first one to have this idea. See Ellen Louise Madison’s University of Rhode Island dissertation. Yes, to not say “first edition” is pedantic. But I think in this case such pedanticism matters because scholars have treated Stowe’s work as what Hershel Parker calls a verbal icon (Note: Parker’s usage is a bit sarcastic: he’s pointing out that what Wimsatt calls a “verbal icon” is based on an inattention to textual transmission.). On the whole, though, I’m pleased that scholars are generally inattentive to the complexities beloved of textual scholars–that “small band,” to borrow Tanselle’s delicious phrase–because they give me a rationale for my work. When I am inattentive to the complexities that fascinate other scholars, I expect them to offer helpful criticism or acidic scorn, depending on their own attitude toward the generosity that should be extended to scholars with different interests than their own
But just because I derive intellectual satisfaction and posit a scholarly career path to examine these topics does not mean that such a study is intrinsically important. Rather, one develops an argument for why such study is important in the present moment, in terms of concerns that animate the profession. But accidents predominate on the road to insight; nonetheless, it is only in relation to key concepts that order an outlook on the world of scholarly discourse that seeming accidents are revealed as significant. A concept that I’ve found quite helpful for defining digital humanities as I practice it is a distinction drawn between book and text. So I will explore that distinction in response to Les’s revision of a statement that he now considers incautious. So let’s take up his revised version: “Where book is used interchangeably with the word text, I will more cautiously assert that the replication of the book (text) should not be the primary goal of the digital humanities scholar.”
I tend not to let my definition of book become overcomplicated, but I do not use it interchangeably with the word text. I will surmise that your concern has to do with the tangible versus intangible nature of such a distinction. I recognize the theoretical bent of those who insist that text is not intangible as a legitimate and philosophically powerful contrast with my definitions. Departing slightly from Peter L. Shillingsburg’s definition of “document,” I consider a book to be tangible, to consist of the physical material, typically paper and ink, bearing a configuration of signs that represent a text, and consisting of more than one sheet gathered or bound in sequence.” The act of gathering and binding makes a book in a conventional sense. A text, again following Shillingsburg, is the “actual order of words and punctuation contained in any one physical form” (46). Like G. Thomas Tanselle, Shillingsburg insists that a text is intangible, has “no substantial or material existence” (46). If a text is intangible and a document is only an index for its reproduction, Jerome J. McGann disagrees with the premise that text and document can be distinguished, preferring instead an approach that is easy to remember if one thinks of it as an erotics, where reproduction is not fully meaningful unless one engages with the ways that the tangible and the intangible get it on in all kinds of ways (with no one on top). I gloss. McGann defines his approach in terms more respectable: “The sonic and visible features of text are, so far as the poets who make these texts are concerned (or the readers who engage them), nearly as apt for expressive poetical purposes as the semantic, syntactic, and rhetorical features. Each of these features represent fields of textual action, and while each considered individually (abstractly) may be described in a hierarchical scheme, the recursive interplay of the fields produces works without a governing hierarchy” (Radiant Textuality, 189). Like I said, an orgy.
I am in agreement with Professor Harrison when he suggests that a discipline which defines itself as digital humanities must have interests other than merely reproducing page images of books. If digital humanities is to form a master discipline–as literary study in U.S. at mid-century pompously imagined itself–it has to be open to multiple approaches, even if the multiplicity of approaches threaten disciplinary boundaries. (Readers of Willard McCarty’s Humanist list will recognize the recurrent outbreaks of such discussions there). But reproduction, engaged critically, is a promising area of research, even if one arrives at critical engagement through accident and luck. I’ll highlight two examples. In the example that I used above, “first edition,” my initial sense was more of selfish interest in my own project. I did not want to call the two-volume Jewett imprint the first edition because most of the National Era printing in fact preceded the book. Why, after all, should I extend the Jewett edition a courtesy that it does not deserve? My resistance to common nomenclature thus became a guerrilla commentary. But I was unable to fully formulate that thought until my advisor questioned my resistance to “first edition” during my oral defense: I had not applied the term consistently in the draft of the dissertation that I submitted for the defense. After he challenged my oral insistence, I went back and changed my wording throughout dissertation (using find and replace). Hereafter, I’m pedantic about not calling Jewett version the first edition of Uncle Tom’s Cabin.
A second example concerns Les’s acknowledgment of my “sound” editorial approach for reproducing the newspaper numbers that lack installments of Stowe’s text. Let’s peek behind the curtain on my project. The Small Special collections digital staff trained me on procedures for reproduction. Their procedures took the physical object as the unit of reproduction. Because the Barrett copy of Uncle Tom’s Cabin was collected in the form that it was, I photographed each newspaper number within the Barrett copy in order. When it came time to plan my own digital project, which included a facsimile reproduction of the Barrett object and a transcription of Stowe’s text, I had multiple choices. But I had already completed a digital reproduction of the Barrett object that included photographs of covers, end papers, and the pages of numbers that lacked installments of Uncle Tom’s Cabin. The “sound” editorial approache, I would submit, is an effort to deal thoughtfully with an institutional procedure of reproduction that contrasted with my own interest in the transcription. Had I been in charge of the reproduction–not forced to engage against institutional practices–I might well have decided to reproduce only those pages that include Stowe’s text. But because of my interaction with a field of institutional practices for image reproduction, I chose to reflect, chose to make a decision about the relationships (which, because I’m writing a dissertation, had to gesture toward theoretical soundness).
I consider the actual choice in part an accident because the object of study would have faced a different choice in many institutional frameworks. Most long runs of the National Era are bound into annual volumes of 52 numbers. At another university, I might well have encountered Uncle Tom’s Cabin as a periodical object that consisted of installments separated into one 1851 and one 1852 annual volume. Because I had plowed through Ralph Hanna’s Pursuing History: Middle English Manuscripts and Their Texts (Stanford: Stanford UP, 1996) as part of my reading list and was familiar with the struggle in Emily Dickinson studies over folio bindings, I had a ready reason (in brief, respect for the object as found) to emphasize the importance of the particular binding of the UVA copy. The Era/Jewett distinction not as exciting to present-day Stowe scholars as Hengwyrt/Ellesmere among Chaucerians or folio/quarto among Shakespeareans, but the enterprise of Stowe scholarship is still in its textual infancy.
While I insist that some part of my editorial decision was a matter of reconciling institutional practices to what I intended to accomplish in an edition of Uncle Tom’s Cabin, I’m more sensitive to books as objects because of the work of McGann and the formative influence of my work at the Blake Archive. Without the pervasive attention to digital objects encouraged by McGann and the mind-boggling attempts to understand objects in the Blake Archive sense, I would be less sensitive to the multiple structures of books when I came to reproduce texts as book objects. In the near future I envision offering an editorial rationale why Stowe’s Key to Uncle Tom’s Cabin is not part of the work. But that decision can only be formulated in relation to actual editorial work, because whether novel and key exist as independent works or side-by-side as parts of one work is a decision that has been made hundreds of times to prepare physical objects. A side note, novel is another word always, in Byrant’s phrase, fluid, always potentially under erasure. I discuss Stowe’s work reluctantly as a novel because her series of sketches precede the institutionalization of novel as the generic word for long fiction. I’m sensitive to the definition of work in reference to Stowe’s fiction only because when examining Jewett paperbacks of UTC I can’t escape the fact that the paperbacks have been frequently bound together with the Key. The two were designed to be bound together, and at least at some level, including in Houghton Mifflin’s Writings edition of Stowe’s works in the late 19th Century, narrative and key are imagined as two parts of the same work.
As we are in dialogue, I will offer not a conclusion but another observation. Students of digital humanities are free to define the discipline such that it is not engaged with the interaction between books as physical objects and texts as computer-based representations. But I find that thinking about these things helps one to re-think what literary works are and what reproductions include or omit. If you reproduce a book without thinking, you are not engaged in digital humanities. But if you think critically about the two distinct forms of representation–if each informs our modes of thought about the other–then such a subject is part of digital humanities (at least when scholars with a background in literary study engage it).
Postscript: I thank Professor Harrison for the first serious reading of my dissertation edition as a digital object.
1) The formal definitions that define analytical bibliography as a discipline were laid out by Fredson Bowers in Principles of Bibliographical Description (1949, see pgs. 37-39) and subsequently refined by G. Thomas Tanselle in “The Bibliographical Concepts of Issue and State” (PBSA 1975, see 18-21).
2) Kirkham, E. Bruce. “The First Editions of Uncle Tom’s Cabin: A Bibliographical Study.” PBSA 65 (1971): 365-82.
3) Scholarly Editing in the Computer Age. Ann Arbor: U of Michigan P, 1996, 47.