Author Archive

Response to Collaborative Reading of Jonathan Swift in Print and Manuscript

August 4, 2010

First, I would like to thank Anna for organizing this reading of my book and all the reviewers for taking the time to read and evaluate it so carefully. This is my first book, and I had expected a two- to three-year wait before reading reviews of it. I certainly did not expect to have encountered a set of seven thoughtful reviews within two months of its publication — one of the many virtues of online collaborative reviews, a subject Eleanor recently discussed. I’ve learned a great deal from these reviews and the comments that followed. In fact, because everyone has raised so many interesting issues, I’m finding it difficult to begin. I will try to address one central issue raised by each reviewer, weaving into my discussion some of the responses to those review posts. I will necessarily have to omit some of the points that have been raised.

Dave Mazella began the discussion by asking the important question of how we talk about eighteenth-century book history: “do we have, or need, an alternative to the unitary term ‘print culture’ in our analysis of the eighteenth century?”. I’m glad to have the chance to address this question, since for the most part I avoided doing so in my book. In fact, I use the term “print culture” only once (205) and then with scare quotes. My avoidance of the term does not signal that I think it has no use in scholarly discussions; rather, I think its value depends greatly on the context of the larger scholarly project in which it is used. Macro-level investigations that chart the major shifts between the pre- and post-Gutenberg eras will depend greatly on an operative term such as “print culture” (obviously I’m thinking of Eisenstein here). But since my book was limited to not just Britain in the first half of the eighteenth century but also to a single author in that milieu, I thought that such a micro-level study would help to show us the limitations and even misconceptions that such macro-level terms imply when used in a more limited disciplinary context.

To put this in more concrete ways: we should not expect that any coherent definition of print culture will encompass such widely different artifacts as a massive folio edition of Clarendon’s History of the Rebellion as well as a broadside poem or single issue of a newspaper. Permanence and ephemerality co-exist here, and the more that a conceptual term embraces paradox and contradiction the less utility it will have for the purposes of definition and argument. In addition, whenever examining any time period up close, one will find overlapping and competing “print cultures” that vary according to cost, format, genre, etc. Perhaps the ultimate lesson is that in such narrowed contexts the term has greater value in the plural form, evidencing a kind of multi-cultural media world.

And to answer Mazella’s question from another direction: it would depend on what else we know of other authors in the period, writing in other genres and not necessarily in metropolitan areas. I had hoped to suggest ways in which Swift was or was not typical of his age in regard to the use of print and manuscript, but I came to recognize that much more needs to be studied before one could begin to address that topic. Studying the examples that Eleanor and Al Coppola mention is one good place to start.

This issue connects to Eleanor’s question about the ability to conduct research in this field given the materials that have or have not survived, especially for non-canonical authors. When I started doing textually and bibliographically related research in graduate school, I had assumed that the major discoveries in this area have already been made and that there was little left to do. But I have come to learn that there is much more to find and many more questions to ask that haven’t been considered yet. In addition, there are now some great research tools for the study of manuscripts in our period, especially for manuscript verse. Eleanor alluded to Alexander Lindsay’s Swift entry for the Index of English Literary Manuscripts, a compendium that is limited to “major” authors from 1500-1900. But there are other resources to consult to learn about major and so-called minor authors from our period. Those interested in further study should consult the following:

I am not as aware of similar resources for non-verse material, but now and then something interesting will turn up while searching Archives Hub.

I would also like to acknowledge Eleanor’s especially perceptive reading of my title.

Ashley Marshall writes thoughtfully about the ways we might revisit Swift’s canon and his attitude toward his own canon, and suggests that Swift’s attitude might change, perhaps in unpredictable ways. I’m greatly interested in such issues as well, even though I mostly avoided dealing with them in my book. (I deliberately avoided addressing works that have a doubtful status in Swift’s canon; such doubtful attributions are receiving renewed scrutiny in the Cambridge Edition of Swift currently being edited and published). Ashley cites two examples: “Cadenus and Vanessa” and “On the Words ‘Brother Protestants and Fellow Christians.'” I think “Cadenus and Vanessa” deserves more investigation about its publishing and textual history, and I haven’t done the necessary research to comment on it at length — except to say that in publishing it in the joint Miscellanies Swift perhaps defused the appeal of those unauthorized collections that contained it.

The example of “On the Words ‘Brother Protestants and Fellow Christians'” shows just how variable a poem’s publication history can be. Ashely is correct that this poem appeared in Faulkner’s 1735 edition; it did so with this prefatory and canny remark: “The following Poem having been printed in London, we have thought proper to insert it here, not doubting but it will be acceptable to our Readers; although we cannot say who is the Author.” (A similar remark prefaced the lampoon “The Virtues of Sid Hamet” in the 1711 Miscellanies.) Bettesworth’s name was fully obscured by asterisks in Faulkner’s text, but the “Sweats-worth” rhyme still gives it away. And it’s worth noting that “On the Words …” was added quite late to Faulkner’s edition, as it appears in the final full gathering of volume 2. So we’re dealing with a situation in which a poem (written while Faulkner was printing the edition) appears in an authorized collection with a note explicitly questioning its canonical status (though with a nudge and a wink). Who exactly was involved in this decision? Did Swift suggest this poem’s inclusion? Or did he pretend he knew nothing about the poem, which shifted the decision onto Faulkner about whether to include it or not? In such situations, Ashley is right to wonder exactly how if at all authorial control is being exercised. Clearly, Faulkner ran a risk in printing the poem, and his boldness was repaid in 1736 when Bettesworth had Faulkner arrested for publishing some relatively mild comments.

A final thought about Swift’s canon: even if Swift did not think he was assembling a canon, some of those in his circle did. Here I’m thinking especially of Charles Ford, who was perhaps preparing an edition of Swift’s works long before Swift was (see the letters between Swift and Ford in 1733). We have here more material for a study of social authorship, a subject I pick up on below.

Randy Robertson rightly emphasizes the need for scholars, perhaps especially those working in seemingly dry subject areas such as textual studies, to craft narratives that engage their readers. And I agree that Robert Darnton is able to entertain his readers while also educating them in a way that I am still trying to master. Perhaps I did a better job with the narrative in Chapter 5 than in Chapter 4, but that is certainly for readers to judge. Often, and especially in the second half of my book, I am simultaneously working with two narratives: the narrative of those writing and publishing in the past and the narrative of reconstructing their activities and the complex textual histories that resulted. At times I found it a challenge to attend to both narratives; in a sense, I had to be true to the information available to us, including the often fragmentary nature of that information, and tell a story that would maintain the reader’s interest.

One issue I struggled with while trying to create appropriate narrative frameworks for these chapters was that I often lacked knowledge about how these texts (whether in print, manuscript, or a combination of both) were received by their readers. In Chapter 6 I attempted a necessarily speculative model about that reception and in Chapter 2 I could draw from readers’ recreations of Swift’s writings, but elsewhere I thought I didn’t even have a basis upon which to construct a speculative model. It seems to me that in order to create a compelling narrative in book history, one needs the perspective of both the producers (writers, publishers, etc.) and the consumers (“regular” readers, the political authorities, etc.). Attending to both will provide a dramatic element to sustain the narrative, and this may be one reason why Darnton’s writings are so enjoyable to read (in addition to his own skill as a writer).

While commenting about “The Legion Club,” Al Coppola wonders about the status of the final couplets added to the poem, shrewdly referring to their inclusion as “a moustache on the Mona Lisa.” I suspect those couplets will always remain a bit of a mystery as to who was responsible for adding them. I gave my arguments for why I believe Swift was responsible, though Coppola is right to explore other possibilities. I wish I knew of some way to solve this puzzle with greater confidence, but I don’t.

Coppola’s remarks gave rise to some interesting comments on the nature of social authorship as it relates to Swift. I’m glad that this topic generated such interest, since we need to do much more to situate works by Swift and other canonical authors in relation to works by lesser-known figures. For my current project, James Woolley and I are editing Swift’s complete poems for the Cambridge edition, and we plan to include the poems that Swift’s poems responded to (Harold Williams had also included such poems) as well as some poems that were part of a verse exchange that Swift’s verse participated in. We hope that doing so will allow readers to examine Swift’s poems in dialogue with those of his contemporaries, including those contemporaries whose identities are not known. We’re also going to pay careful attention to canonical issues and the extent to which we can attribute specific poems to Swift with varying levels of confidence and uncertainty. James has written of the need to free ourselves from “the tyranny of the individual talent” and take seriously the possibility and even likelihood that one of Swift’s contemporaries may have written a poem otherwise attributed to him (“The Canon of Swift’s Poems: The Case of ‘An Apology to the Lady Carteret,'” in Reading Swift: Papers from the Second Munster Symposium on Jonathan Swift, p. 246). In a brief entry on “Attribution” for the recent Oxford Companion to the Book, I referred to “the centripetal force of the major author” to describe the power that a major figure exerts on contemporary works of uncertain authorship. I also wrote that “scholars who offer claims of attribution would do well to remember that the vast majority of works are written by minor authors, including authors whose identities have long vanished from historical memory.”

A recurring phenomenon in studying historical reader response is the way in which readers adapt and use an author’s works for their own purposes, whether or not such purposes align with those of the author. That phenomenon is no less operative among readers of scholarly works today. I write this in part to justify David Brewer’s interest in taking my research on a ride of his choosing, while graciously welcoming me aboard. I would be very interested in knowing with greater clarity just which printed texts prompted readers to take up pen and annotate. Then as now I suspect some readers did not want to “mar” their personal copies; others perhaps couldn’t help doing so. It’s worth noting that for some time collectors (and thus libraries) viewed such annotated copies as less desirable, a trend not followed by collectors such as James Osborn, as can be seen by viewing the many items he collected now at the Beinecke Library. A brief research visit to the Beinecke will remind us how much more work there is to do with eighteenth-century manuscript materials (including annotated print copies). So one wonders if the surviving evidence in libraries is to some extent biased in a way that obscures the act of annotation. Interested readers might consult H. J. Jackson’s Marginalia: Readers Writing in Books, and (I hope) work by David in the near future.

Are there any copies of Tristram Shandy that in fact contain a sketch of the Widow Wadman on the blank page? I believe that David has researched this topic.

I’m also grateful to David for alerting me to yet another annotated copy of Swift’s “Verses.” James and I are always interested in learning about these. The ones that have escaped our attention are likely to be in libraries with good general collections but too rarely frequented by eighteenth-century specialists.

I’m somewhat relieved that David’s discovery doesn’t dramatically alter my argument, though as Anna notes, that is a possibility and one that I expect to happen at some point. I’m very grateful to her for making this truly collaborative reading possible. I’ve learned much from each of you, and want to thank you again for devoting part of your summer to my book.

Steve

Advertisements

Digital Textbases and Optical Character Recognition (OCR)

July 16, 2009

Experienced users of ECCO know about the limits of its full-text capability. The long s in eighteenth-century fonts is one of many peculiarities that can wreck an automated effort at optical character recognition (OCR). Though I’m grateful that I can search ECCO and other databases using full text, I often wonder how complete my search is. I usually get a sense of how many false hits I find, but how many true hits am I missing? How accurate are the full-text capabilities of these resources?

A recent article presents a method for assessing the accuracy of OCR using the British Library’s 19th Century Newspaper Project as a case study:

Simon Tanner, Trevor Muñoz, and Pich Hemy Ros, “Measuring Mass Text Digitization Quality and Usefulness: Lessons Learned from Assessing the OCR Accuracy of the British Library’s 19th Century Online Newspaper Archive,” D-Lib Magazine 15.7/8 (2009).

This is available at:

http://www.dlib.org/dlib/july09/munoz/07munoz.html

The article briefly mentions Gale’s Burney newspapers project. One of the good points in this article concerns how we should measure accuracy:

Given a newspaper page of 1,000 words with 5,000 characters if the OCR engine yields a result of 90% character accuracy, this equals 500 incorrect characters. However, looked at in word terms this might convert to a maximum of 900 correct words (90% word accuracy) or a minimum of 500 correct words (50% word accuracy), assuming for this example an average word length of 5 characters. The reality is somewhere in between and probably more at the higher extent than the lower. The fact is: character accuracy of itself does not tell us word accuracy nor does it tell us the usefulness of the text output. Depending on the number of “significant words” rendered correctly, the search results could still be almost 100% or near zero with 90% character accuracy.

The term “significant words” refers to words that users are likely to search for, in contrast to function words (pronouns, prepositions, etc.). A textbase’s accuracy in terms of “significant words” is an appropriate yardstick for how useful its full-text search is.

The full article merits reading. The authors found that for significant word accuracy, the 19th Century Newspaper Project was 68.4% accurate and the Burney Newspapers was 48.4% accurate. Eighteenth-century newspapers can be astonishingly difficult to read even in the originals, so this low percentage is not that surprising. I suspect that ECCO is somewhere in between these two percentages.