Copyright notice: all material on these pages is derived from sources in the public domain.
My approach to hypertext/online editing is indebted to Jack Lynch.
The text of this version of A Tale Of A Tub has been scanned from the 1850 edition of Swift (The Works of Jonathan Swift, with Memoir of the Author, by Thomas Roscoe. London: Henry G. Bohn, 1850), edited by Thomas Roscoe. The scanning was done by myself in the `Eaton' computer lab at Tufts University in April, 1996. The scanner used was a Hewlett Packard ScanJet IIcx, and the OCR software used was Omnipage Pro.
Though the Roscoe edition generally conforms to the 1710 (standard) edition of A Tale of a Tub, it is not precisely my own. I have used it because it is the oldest edition to which I have had sustained access to (i.e., the oldest edition I was allowed to take out of the library).
The definitive 20th century edition of A Tale Of A Tub is the one edited by A.C. Guthkelch and D. Nichol Smith (A Tale Of A Tub, To Which is Added The Battle Of The Books and the Mechanical Operation of the Spirit. Oxford: Clarendon Press, 1958.). Guthkelch and Smith make every effort to precisely replicate the form and content of the 1710 edition, including the two separate essays which were appended to A Tale Of A Tub. Since I have not had access to an original edition of Swift, or even a facsimile thereof, I am trusting Guthkelch and Smith as reference: when I refer to `the 1710 edition,' I am referring to Guthkelch and Smith's version of the 1710 edition.
It may seem remarkable that I would have the audacity to publish an edition of a text without seeing a single physical copy of an edition published in the author's own life-time. In some ways (this might be a predictably shameless poststructuralist disavowal of responsibility...), it may be only appropriate that the primary reference of a hypertext edition of a text would itself be able composed without reference to a physical `original.' Indeed, it would be impossible to make the experience of reading A Tale Of A Tub online anything like what it might be like to read an original edition. Specifically, there's no danger that this edition will decay or disintegrate. `Original' editions need to be preserved and reserved; only a limited number of highly qualified hands can be allowed to touch them. In comparison, an online edition is absolutely free and effectively permanent (is it therefore less valuable?). I imagine that this text will be read by those scholars and enthusiasts who are awash in the discursive and material productions of the 18th century: those for whom discourse should have no limits.
Perhaps a stronger defense of my editorship is the following: contemporary readers of Swift have never seen the `original' text either. They have access to it through either the Norton Edition, the Oxford Edition, or the Riverside Edition, all of which have been `edited,' and in some ways, altered, in the course of their editings.
An online editor, one not constrained either by the profit-motive or by the constraints of paper, might be useful in the following ways. First, it's an opportunity to intervene with and challenge the effects of three-hundred years of battles for editorial authority. For instance, why has every recent standard edition of Swift incorporated the "Hawkesworth" footnotes into the body of the text, while relegating the editor's own footnotes to the `back of the book'? What makes Hawkesworth's commentary so worth passing on? (In many cases, it isn't helpful; in some cases, it unwittingly participates in the `Modern' bewilderment that is precisely what Swift is satirizing.) To avoid passivity in this regard, I have tried to clearly differentiate Swift's footnotes from my own, and to indicate that any other footnotes I have deemed to include (at present) are also my own doing. In most cases the `other' footnotes I have used have come from Guthkelch & Smith; they are present in my text as scholarly citation.
Second, an online edition might direct the reader to the actual texts (that is, the hypertexts) that an author might refer to, rather than summarily and reductively `explaining' or `interpreting' an author for the benefit of instant readerly legibility. The hypertext editor need not be the `meaning-controller' that a paper-editor in many cases has to be. Unfortunately, many of the texts it might have been suitable to refer to are not yet present online, or I have not yet found the relevant links. All suggestions to this effect would be appreciated. A specifically useful project for the immediate future would be the fragment of William Wotton's Reflections Upon Ancient and Modern Learning which deals with A Tale Of A Tub.
The 19th century is the Dark Ages (perhaps- the Cold War) of editorial discretion. The primary motivation in the various 19th century editions I've seen seems to have been quantity (the mass-production of the most extensively reduced texts possible) rather than accuracy, or the comfort or sanity of the reader. The Roscoe edition, which has a fifty- page biography of Swift, lacks even the barest of editorial explanation of its presentation of A Tale Of A Tub.
I have made a number of changes to my original scan of the Roscoe Edition, some of which have been to replace footnotes from the 1710 edition that the Roscoe edition omits. In other cases I have removed footnotes that were added in by editors such as Sir Walter Scott or John Hawkesworth, which functioned mainly to explicate some of Swift's more obscure references. My personal reading of Swift inflects such a decision: I don't think it's important to know all of the references in A Tale Of A Tub to be able to follow rather closely the `substance' of his satire.
Roscoe makes a number (hundreds, actually) of spelling changes from the 1710 edition. I have only reverted a few of them. It simply didn't occur to me until late in the process that it was an important issue. We do have the OED to indicate how certain words might have been used in Swift's time, to consider what the resonance of a particular spelling might have been in the early 18th century. But without instant access to an OED or comparable etymological/morphological reference source online, the project of preserving Swift's original spelling would likely only to further decrease the legibility of the text. In a different vein, spelling does not appear to have been as standardized in Swift's time as it is now. The `great confinement' of stray letters was still in process. I confess I've simply dodged the problem: my edition is a `spell-checked' text; most spelling has been set to conform to contemporary British-English standards.
In certain cases, Swift's spelling and grammar are from Pluto (or... Uranus?). Swift uses many neologisms, extremely obscure usages, and even a few completely unintelligible non-words in A Tale Of A Tub. Further, in the 1710 edition, the presence of translations of `Latinisms' are inconsistent (as Wotton points out, the quote from Iranaeus on the cover is "pure gibberish"). For me, the obscurity is part of the satire: only the most obsessive `Modern' would bother to track down and compare his quotes from obscure Romans. If a simulation of the effect of the reading-experience of A Tale Of A Tub is the desired function, the Latinisms continue to belong in the text, quite possibly without annotation.
To conclude: in the course of this series of observations on the editorial policies I have followed in producing A Tale Of A Tub, it might seem as if I have programmed the text to be used in fairly specific ways. That is true. My choice of textual authorities (i.e., Guthkelch & Smith over Scott, Hawkesworth, Davis, or Roscoe), my modernization of spelling, even the language I'm using in this very essay: all of these elements allow some modes of reading and disallow others. I don't think my edition of the work is overly restrictive in terms of its potential readership, or at least I hope it's no more restrictive than any edition.