Using a passage from Chaucer’s Monk’s Tale, this paper closely reads networks of readers and scholars through the lens of efforts toward exact transcriptions. It examines paratexts, apparatuses, methodologies, and labor as emblems of the underlying politics of literary reproduction and the difficulties of exactness. I argue that closely reading previous editor’s efforts toward exactness crucially informs digital reproduction. Thus, the historical materials examined are all editions of Chaucer where the author made some effort at exactness. The different approaches are unpacked into their functions of aesthetics, semiotics, and comparison. The paper concludes considering these functions as enabling a discussion of the possibility of best-practices for reading and creation in the Age of the Digital Reproduction.
Cousins of Deeds and Tricks of Documents:
Diplomatic Transcription and Shifting Senses of Exactness Toward the Ecosystem of Digital Reproduction
1> Edited editions were the sine quo non for literary studies of Chaucer when Atlantic steamer tickets were expensive, photocopies were a fantasy, and digital cameras were relegated to science fiction. To study such a canonical author, scholars would depend on transcriptions, and for deeper study of an author- diplomatic transcriptions that were accurate to the glyphic features of the text. Even if budgets allowed access to the over eighty original manuscripts of Chaucer, the author’s transcription still mediated between the reading room of a great library and their book-lined study where they wrote. This separation privileged the work of great scholars, who had the reputation to command budgets, teams, and infrastructure. Today, however, digital reproductive networks purport to offer an equalizing road to scholarship. The harried scholar, attempting to squeeze more from the fountainhead of English poetry, can outsource their labor to distributed networks of readers and familiar spirits, who by digitizing materials have begun the scholar’s work already. Yet, these readers bring their own biases, and the network topology privileges certain patterns. We will attempt to read the networks and methods in this paper as a way of demonstrating modern critical methods.
2> Bad editors of Chaucer are well known and plentiful. John Urry’s[1]example of altering words to maintain an imagined meter, which is discussed more below, is a particularly egregious case of changing the content to make it more exact, yet making it inaccurate. Of course, this is still an effort toward exactness using the technology of Urry’s time. Digital tools simply represent a new station on the way of precise reproduction and so to place them one must first consider earlier specialized efforts toward exactness. So, while it is tempting to begin by describing Unicode, the Medieval Unicode Font Initiative, and how to use technologies with text encoding, beginning with digital technologies is a positivist totalizing gesture implying that the new digital systems supersede centuries of intelligent people aiming for exactness. This gesture may trap us in the same way that Urry was trapped by the thinking of his time. The digital is but a small and recent innovation in the long history of textual editing, which must be considered in full course. One of the oldest tools for efforts toward exactness is the diplomatic transcription.
3> A diplomatic transcription has always been considered somewhat an odd duck and has only become more so in the twenty-first century. The sense of ‘diplomatic’ as something related to an exact copying of documents, usually legal but occasionally literary, seems to originate from Gottfried Leibniz’s effort to write a history of the House of Brunswick that contributed to the volume Codex Juris Gentium Diplomaticus, of 1693. I suspect the word may have been in use prior to 1693, but the relation to the word “document” via way of “diploma” makes this derivation plausible.[2] The method of reproducing the content for a particular intention is what interests this paper, which Leibniz illustrates nicely.
4> Leibniz’s near-earliest instance of diplomatic transcription deals with accurately reporting the contents of legal documents. He was attempting to create an exact copy to capture the aura of the original copy, but of course an exact copy, especially in 1693, is something of an impossibility. Every purported instance of duplication is an act of reproduction with the idiosyncratic act of writing filtered through the equally idiosyncratic reading-act. What makes considering reproductive efforts aiming at exactness interesting are the paratextual apparatus that accompany them; some writers may think their copy is exact if they cannot read certain aspect of the original text, or they may simply be striving for exactness within the limitations of their technology. In Leibniz’s case, the technology was a new typeface cut specifically to make more accurate reproductions, but his techniques for diplomatic transcriptions were reused and developed into their own language of reproduction. The effort of exact copy morphs into its own discipline of diplomatic-type transcriptions and we have a new discipline of similar, exact, diplomatic, quasi-facsimile, and other transcriptions serving different purposes.
5> Not only is the truly exact impossible, but even efforts toward it may not reward the significant labor required. Scholar’s lack of time explains some of the rarity of diplomatic transcriptions, but begs a question in the twenty-first century where digital cameras are common, powerful, and inexpensive. Can we achieve something closer to exactness at less cost? What, if anything, is a diplomatic transcription--and similar efforts at exactness--good for when we can take a picture? How does the changing technological environment inform digitization best practices? When is similarity good enough?
6> This essay attempts to answer these questions by looking closely at a tiny piece of Chaucer’s work, the beginning of the Monk’s Tale within the larger Canterbury Tales and how its representations have changed over the long history of Chaucer scholarship and these representations participate in creating meaning. The particular tale was selected because it has been widely reproduced over its roughly six century history and so possesses a number of textual complications and has a variety of paratextual features in different editions. It thus provides something of a emblem for the general problem of the exact, similar, or diplomatic reproduction. We will explore efforts towards exactness pragmatically and historically by elucidating potential purposes and intentions. The historical-intention approach ultimately exposes certain common aspects of efforts to exact copy which can be applied in different situations. Ultimately, the historical reading will be applied to suggest new editorial approaches, leveraging current digital technology. While the end result looks like a list of best practices, the process of discovering practices is itself useful for understanding humanistic attitudes toward knowledge, reproduction, text, reading, and writing. This essay therefore codes an experiment in speculative computing, which emphasizes aesthesis and its non-totalized, subjective, knowledge, over mathesis’s effort to totalize knowledge into formalized logic.[3] Diplomatic transcriptions and digital facsimiles provide complicated sets of tools for editorial work that must be closely analyzed to be used well in producing efforts toward exact copy.
The Aims of the Exact Copy
7> Exact copy is something of a bugbear- we allude to it to indicate intention and mechanism, but ultimately the idea is unreachable. Examining attempts, however, will give us a better sense as to why someone might engage in diplomatic, or digital, reproduction. F.J. Furnivall’s Six Text[4]version of Chaucer neatly illustrates the three core purposes of attempting exact copy (see figure 1).
Fig. 1
Caption: Three columns of F.J. Furnivall’s Six Text edition of the Canterbury Tales.
8> Visually, the typographical color is varied and pleasingly patterned. The virgules form a pattern of lines, particularly in the Ellesmere facsimile,[5]that recapitulates the weaving right edge of the poem, led by multi-line initial letters including a large ‘I’ that seems to anchor the identity of the speaker as the source. This visual aspect is what I deem the aesthetic/bibliophilic purpose of efforts at exactness. Looking more closely, we see variant glyphs such as the ‘h’ with an otiose slash, the ‘r’ with a hook, italic expansions of abbreviations, thorns, and yoghs. These signs are both symbols of particular editorial decisions and icons connecting to larger cultural works. They signify, when unexpanded or otiose--unexpandable--abbreviations, a cluster of possible readings each of which is a possible expansion. At the same time, they symbolize the cultural code of scribal practice. These symbolic aspects are what I call the semiotic/semantic purpose of efforts toward exactness. However, looking across the three columns reproduced in figure 1, we see similarities and differences that indicate something about both the artifact they refer to and the ideal copy,[6]perhaps, lurking behind the physical artifacts. The effort to collate is what I call the comparison purpose of efforts toward exactness. We thus see three purposes represented here- the aesthetic/bibliophilic, the semiotic/semantic, and that of comparison.
Aesthetic/Bibliophilic Purposes
9> The aesthetic/bibliophilic analysis of texts has been introduced by various authors, so its inclusion as a way of understanding efforts toward exact copies should come as no surprise.[7] The reading experience is generally visual, so reproducing the appearance of a text carries through some of the paratextual apparatus.[8] Aesthetic considerations are not always attached to an effort toward exact copy, but in the case of Chaucer we have a tradition of particular typefaces. Consider the appearance of the Monk’s Tale from the Kelmscott Chaucer, (figure 2.)[9]
Fig. 2
Caption: The beginning of the Monk’s Tale from the Kelmscott Chaucer.
10> William Morris designed the Chaucer type seen in this image specifically for Chaucer’s works, continuing a tradition of using Blackletter typefaces for the Monk’s Tale. This Blackletter tradition alludes to the earliest printed versions by Caxton[10](done in what looks very much like anglicana formata, Blackletter), Thynne’s edition[11](in semiquadrata, Blackletter), and the manuscript tradition[12](which was anglicana formata in certain notable cases).
11> The use of Blackletter, sometimes called Gothic, resonates with a long tradition of allusive typography, particularly associated with the distinctive Blackletter family.[13] The aesthetic tradition of Blackletter was noted by many readers. In 1781, Horace Walpole wrote to William Mason, "I am too, though a Goth, so modern a Goth that I hate the black letter, and I love Chaucer better in Dryden and Baskerville than in his own language and dress" continuing to suggest the use of the Blackletter carried some visual meaning. Yet, some who were aware of the tradition of Blackletter were not so effusive. As late as 1841 Isaac D'Israeli said that "Readers will be appalled by having to face a massive tome dark with the Gothic type, whose obsolete words and difficult phrases, and for us, uncadenced metre, arte to be conned by a glossary as obsolete as the text, to be perpetually referred to, to the interruption of all poetry and all patience.” Richard Hengist Horne in 1841, regarding his modernized Chaucer said "with the true but narrow devotion of the best men on the black-letter side, and their resistance to all attempts to melt the obsolete language and form it into modern moulds ... the Homer of English poetry continues unread except by very few," tying the idea of Blackletter to unreadability and loss of appreciation for English literature.[14] While differing on the effect of typeface on the reader’s experience, each of these readers demonstrates an awareness of the role of typeface as a paratextual apparatus.
12> The Kelmscott Chaucer depicts the fraught meaning of the Blackletter typeface. Aesthetically in the typeface, as well as through the layout, the Kelmscott Chaucer attempts to copy the feel of early reproductions. Morris was alternatively criticized and applauded for his neo-Medievalism; truly, while an aesthetic response exists, the value is fraught. This sort of aesthetic copying occurs in many of the reproductions reviewed in the usage of headings, large initials, and especially decorative images. (the Ellesmere, the 2nd edition of Caxton, the Speght,[15]and later versions of Thynne, all give similar images of the Monk astride a horse.) Yet, the repetition of typefaces and images suggests efforts toward aesthetic exactness which is the easiest purpose for the exact copy to apprehend.
13> Digital transcription systems, of course, have a wide range of typefaces to choose from. The image of the writer procrastinating by changing typefaces thus indicates her attempt at finding the right aesthetic resonances. A way to understand this is Tanselle’s “physical text,” which is the order of the letters and words, independent of layout and aesthetics.[16] The modern digital environment operates on the physical text level, as a string of characters, that can be reformatted with new typefaces, new margins, changing columns, etc. This allows digital transcriptions to be adaptable in the repetition of the graphic aesthetic forms.
14> The repetition of the graphic aesthetic forms extend from the manuscript tradition as well. The Hengwrt and Ellesmere manuscripts bear a striking resemblance to each other, leading some to suggest that they might have been created by the same scribe. They are both in a steady anglicana formata, use similar scribal abbreviations, and have nearly the same catalog of glyphs. They both use a somewhat uncommon abbreviation consisting of a loop on the long anglicana ‘r’ of “hir aduersitee,” but their use of the sigma form of ‘s’ differs. (See figures 3 and 4.)
Fig. 3
Caption: The beginning of the Monk’s Tale from the “Ellesmere Chaucer.”
Fig. 4
Caption: The beginning of the Monk’s Tale from the “Hengwrt Chaucer.”
15> The phrase “blind prosperitee” is also expanded differently in the two versions, which supports the general assumption that either these two texts share a parent or are copied from one another. The transcription gives an example of an aesthetic effort at exact copy and, since the two versions are expanded differently, provides a case where exact copies of the two manuscripts would differ.
Semiotic/Semantic Purposes
16> Similarity in aesthetics in typefaces is also an icon that connects to similar usages through cultural codes and so affects the reading and reception of the work. For example, Blackletter is associated with the Bible and other sacred works as well as German language works.[17] We can read this association as aesthetic, but also semiotic by considering the signification first, although the two are closely related.
17> D.C. Greetham discusses the application of semiotic analysis to texts through the work of C.S. Peirce and Roland Barthes as it applies to critical editions.[18] Greetham essentially posits that second-order systems of editorial apparatuses can be identified as acting on the first-order system of signs that compose the language. That is, an edited edition is distinguished by the presence of two layers of signs- the first-order is normal language, tropes, symbols, codes, etc. while the second-order is the specialized choices of the editor, diplomatist, or other actor in the reproduction of a text. The second-order writes the ‘myth’ that Barthes speaks of, or the digital version of the cultural code of Lotman, which is equivalent to the semiosis (recovering signifying relationships of the text and its history) of the critical edition.[19]
18> The usage of particular types of symbols or culturally-weighted icons allows the diplomatist, or others making effort toward exact copies, to expound on the semiotic structure of the text, but in a partially interpretive way. Diplomatic efforts don’t typically include any strong misprision, let alone weaker forms, but it’s certainly not out of the question for interpretation to overrun the materials. This semiotic expounding is a key difference from the digital image, which slavishly copies the surface and is an act of mimesis (imitative representation of reality) rather than semiosis.[20] The choice of sigla and glyphs allows the diplomatists to communicate subtle information about the artifact that the image does not capture.
19> Note that the transcription of the “Hengwrt Chaucer” in figure 1 expresses the marginal insertion of Adam’s Tale (see figure 4) as a continuous text, but indicates the difference in the apparatus. The Hengwrt manuscript itself does not explain the nature or authority of the note and so the structure of Furnivall’s edition expresses a particular second-order relationship to the underlying text. The editor expresses a judgment about the physical text with the tools of semiotics, which is a technique afforded only in transcription, not in slavish copies. Thus the purpose of an exact copy can be to translate, or even modernize by changing from difficult-to-read handwriting to electronic typography that contains the same graphical elements, the semiotic structure of the original.
20> An example of slavish copy can be found in two editions of Chaucer examined for the current paper study. Skeat’s[21]edition and Ruggiers’s[22]edition capitalize on their respective contemporary technology to attempt an exact copy of the text. Skeat presents a collotype facsimile of the whole work as a mimetic index of the photographic trace, mediated through printing, of an actual writing event embodied in an artifact. While the edition is beautiful and fascinating, it provides nothing but a reflection of the original manuscript. Ruggiers’s edition provides a facsimile combined with a transcription and variorum. Unlike Skeat, Ruggier expands his interpretation with italic expansions and notes, but the semiotics of the exactness of Ruggier’s transcription are hidden in the italic expansions and notes. However, to the extent which he knew the icons, they have been retrieved and represented in the notes. Thus in Ruggiers’s edition we are faced with a plural marriage of two manuscripts, an apparatus, and a bibliography of scholarship. The exactness in Ruggier’s document is both in content and the description of the state of knowledge at the time of its creation.
21> The seeming originals of Chaucer are numerous, or few, depending on how you view the ontology of the text. Manly and Rickert[23]examined the eighty-three known manuscripts in their unparalleled, massive, study of the text of the Canterbury Tales, while F.J. Furnivall looked at six copies simultaneously. William Thynne bequeathed twenty-five Chaucer manuscripts in the late sixteenth century to his son Francis, who was frustrated in his attempt to create a new Chaucer edition by Speght who released his edition first. Francis simply used his collection to write a critical introduction to Speght’s volume.[24] The introduction implies a recognition that possessing a large number of manuscripts bestows apparent authority and interpretive power, even if someone else beat you to the chase in doing and publishing the actual work of interpretation; the ability to compare a large number of texts privileges an interpretation. The privilege of the true text can be found within another purpose of exact, or diplomatic, transcriptions, that of comparison, between near and far manuscripts as well as comparison between presently-known manuscripts and future discoveries.
The Purposes of Comparison
22> Comparison can be seen throughout much of the history of attempts to edit Chaucer. While Francis Thynne personally owned all twenty-five manuscripts he used for comparison, and so could presumably spread them out on a table, neither Caxton nor Tyrwhitt[25]had such a privilege. Caxton famously set his first edition of the Canterbury Tales from a manuscript in his possession, but soon afterwards a purchaser of the first edition paid him a visit and pointed out that Caxton’s printing had errors because his manuscript was bad. The visitor’s father possessed a superior manuscript which he begrudgingly lent Caxton. “He sayd he knew a book whyche hys fader had and moche louyd / that was very trewe / and accordying vnto hys owen first book by hym made / and sayd more yf I wold enprynte it agayn he wold gete me the same book for a copye.”[26] So, Caxton’s earlier printing serves as a comparison with the later manuscript. The comparison between transcriptions of manuscripts was undoubtedly repeated by Tyrwhitt who strove “to give the text of the Canterbury Tales as correct as the Mss. within the reach of the Editor would enable him to make it,”[27]but who had to rely on manuscript copies found in institutional collections and Caxton’s editions. Thus, in these editions, a diplomatic-type transcription enables the exploration of the ideal text or polytexts.[28]
23> Manly and Rickert approached the task of comparison on a massive scale, mediated by a new technology, the photostat. Their team collected eighty-three photostat, or photographic, copies of each of the known Canterbury Tales manuscripts. While Manly and Rickert left their original photostats on deposit at the British Library, their final product has the mass-produced technological feel of their process. The text is a photolithographic reproduction of carefully typed pages of text, with carefully typed pages of notes; these mass-produced products neither honor aesthetics nor the semiotics of the originals. Each character is moderated for maximal efficiency by a typewriter designer and appear as an icons for the normal office work of the day. The work of the semiotic structure is entirely hidden in the British Library archives and indicated in Manly and Rickert’s notes. Aesthetics are dismembered from the text and discussed in separate opening essays. The work of Manly and Rickert in a sense sanitizes the rich experience of manuscript sources by developing an ideal reading, based on scientific principles, and then separating the messiness in a series of notes. The form recapitulates the empirical sterilization of knowledge of the double blind experiment, which in a laboratory becomes something to code and classify. Manly and Rickert approach an exact copy, made through exact copies of distant manuscripts, but with the empirical approach of scientists. Aesthetics, essentially, are lost.
24> Each of these previous examples focus on comparison between presently-known, collocated manuscripts. Caxton and Thynne compared texts at hand; Tyrwhitt compared distant texts without reproductive technology; and Manly and Rickert compared distant texts using mechanical reproduction. Since each of these editors dealt with a finite collection of known texts, their apparatus was hidden or submerged. The raw materials for comparison were unexaminable since once the creators compared them, the exact copies were no longer needed. A contrasting purpose is to compare presently known manuscripts with future discoveries, in which the versions used for comparison must be preserved in some way. A good diplomatic transcription can serve in lieu of access to a manuscript or printed edition that has been lost or destroyed. This role of proxy is suggested by Tyrwhitt when he considers Caxton’s editions as manuscripts; since they represented versions now lost to time, he judged Caxton’s editorial process as sound.[29] The reuse of historical exact copies seems to be the tacit goal of Furnivall when he presented in great detail the diplomatic transcriptions of his six best versions of the Canterbury Tales.
25> Yet, thinking of future manuscript discoveries alone glosses over the connotation of “presently known,” which can also be referring to the limits of current knowledge. A hugely important work in Chaucer scholarship is F.J. Child’s “Observations on the Language of Chaucer and Gower.”[30] Child rediscovers the pronunciation and purpose of the terminal ‘e’ in Chaucer’s language, which had previously been assumed to be simple caprice or flawed transcription. This discovery signaled an end to a type of Chaucer scholarship where an editor would alter the text to make it fit with various poetic ideals. Prior to this discovery some editors would try to fix variants that they saw as errors. John Urry’s edition[31]was notoriously bad in false corrections. His editorial process is described:
His chief busines was to make the Text more correct and compleat than before. He found it was the opinion of some learned Men that Chaucer's Verses originally consisted of an equal number of Feet; and he himself was perswaded that Chaucer made them exact Metre, and therefore he proposed in this Edition to restore him (to use his own Expression) to his feet again, ... That whenever he could by no other way help a Verse to a Foot, which he was perswaded it had when it came from the Maker's hands, but lost by the Ignorance of Transcribers, or Negligence of Printers, he made no scruple to supply it with some Word or Syllable that Serv'd for an Expletive ... Besides all those Methods for lengthening out Verses, which was the thing generally wanted, he had occasion sometimes to shorten them; and for that purpose he frequently makes use of an Apostrophe where he would have a Syllable dropt in the pronouncing of a word, as Sov'rane prize, ...[32]
26> Thus, in Urry’s work we have a sort of anti-diplomatic transcription which attempts to repair errors that were simply a gap in the current understanding of Chaucer’s language. Urry’s edition is also the first to break the aesthetic copying by using a Roman typeface. (See figure 5.)
Fig. 5
Caption: The beginning of the Monk’s Tale from Urry’s edition.
27> His re-interpretive work changed the reading of the Monk’s Tale. Examining line three of Adam’s Tale, we can see that Urry uses the word “strene,” meaning gain or sexual intercourse, rather than the more common “sperme.” The line thus reads “And not begotin of mannes strene [cf. sperme] unclene,” which suggests that the sex act or man’s gain, which could be interpreted variously, is unclean, rather than suggesting that sperm is unclean. Manly and Rickert do not find this reading in any of the manuscripts they review, suggesting that it was one of Urry’s restorations. These restorations earned Urry the scorn of many editors and is succinctly summarized by Tyrwhitt:
The strange license, in which Mr. Urry appears to have indulged himself, of lengthening and shortening Chaucer's words according to his own fancy, and of even adding words of his own, without giving his readers the least notice, has made the text of Chaucer in his Edition by far the worst that was ever published.
He goes on:
I will just add, for the sake of those who may be disposed to make use of this Glossary in reading the works of Chaucer not contained in this edition, that it will be found to be almost equally well adapted to every edition of those works, except Mr. Urry's. Mr. Urry's edition should never be opened by anyone for the purpose of reading Chaucer.[33]
28> It is does not seem coincidental that Tyrwhitt’s criticism prefigured Child’s essay which gave part of the impetus for Furnivall’s diplomatic transcriptions; it seems that there was a growing awareness of the limitations of knowledge and the role that diplomatic transcription could play. We see that a diplomatic transcription, or other exact copy, can serve as a hedge against the current limitations of knowledge, and compared to interpretive reconstruction, diplomatics is more honest and conservative. Diplomatic efforts toward exact copies facilitate intertextual connections with near and distant texts, as well as across time.
29> Explicit intertextual connection are one of the strong suits of digital reproduction. Landow famously goes as far as to suggest that hypertext--by which he conflates most forms of digital reproduction--is the reification of critical theory.[34] The simple hyperlink makes intertextuality tangible for even the undergraduate and properly coded texts can be reused to represent all manner of connection. Had Urry had access to a digital system that could represent uncertainty then he could have indicated the nature of his modifications, and not falsified his own work with the discovery of new pronunciation methods. This, however, is something of a digital-philic positivist argument; Urry did have access to a system that could represent uncertainty in the form of footnotes and the traditional use of italics to represent expansions, he simply didn’t make use of them. So, while the digital offers us new tools for editing, these tools are simply digital versions of what were often more cumbersome tools.
Intertextuality as Key to Diplomatics
30> No matter what one thinks about the validity of the ideal copy as editorial model, or the polytext model, these machinations about exact copies are really about intertextuality: between some ideal and the residue, between residues, or between later reproductions that have no ideal. Aesthetics represent an intertextual relationship with other exemplars of form, while the semiotic are intertextual relationships through icons and symbols, and comparison is explicitly intertextual and can be used to contain the other two. Thus, the purpose of a diplomatic transcription, and by extension, new efforts toward exact copy--digital facsimiles among them--is a close reading of intertextual relationships. The intertextual purpose implies that mindless digitization is not a replacement for diplomatic transcription but a new tool to be used in the aid of superior editing. The key to forming an exact copy is that an editor untangles complicated relationships and reproduces those in the appropriate technological environment. So, our digital camera represents, not a replacement for diplomatics, but a new type of typewriter, that--like Manly and Rickert’s photostat machine--can be used in building the successor to diplomatic transcriptions. We can begin to explore new techniques using the promising digital tools available and therefore begin to approach an understanding of what new diplomatics might be.
Digital Tools
31> Work in new diplomatics, as a subset of digital humanities, can be understood as the building of tools, while the editorial impetus remains firmly grounded in traditional scholarship. From the preceding discussion of diplomatic transcription, we can identify concrete goals to 1) provide a surrogate that brings the original to mind across space (comparison across space and semiotics); 2) provide a structure that can be incorporated into future scholarship (comparison across time and semantics); and 3) create pedagogical ladders (aesthetic/bibliophilic). The first is crucial because the critical work must somehow prove that it encompasses the original. The critical work need not exceed the detail and connections of the original given the limits of current knowledge, however, it should at least “bring it to mind” for those who have viewed the work and not obscure the rich structures. The second goal, providing reusable components, is one which helps further scholarship. It recognizes that current scholarship is beholden to the past and that future scholarship must examine the present. Scholarship which does not generate a new, re-usable, narrative is decadent and degenerate because it would eventually consume all the old narratives and the field would die. The third goal is essentially the goal of all criticism. Were everyone a perfectly informed reader, there would be no need to criticize or to edit. One could be satisfied with the original item and high-quality, although slavish, facsimiles. Recognizing that literature is a discipline implies that there is something to learn, so textual works ought to assist. Possible tools for these techniques are discussed below.
Bring the Original to Mind
32> The quickest, and most widely understood twenty-first technology for surrogation is the digital facsimile. Through hypertextual communication networks (i.e. the internet), scholars can retrieve images, request copies, or make appointments to create their own photographic work. These facsimiles provide the immediately available surrogate that captures the layout. However, diplomatic transcription provides a slightly different surrogate.
33> Handwriting is simultaneously unique and references a class of glyphs that represents some ideal form. The digital, or original, manuscript challenges the reader to learn the idiosyncrasies of the scribe who wrote it, but once the scribe’s hand is learned, it can be read easily. Given time away from the hand one may find it challenging again. The diplomatic transcription aims to alleviate the challenge of paleography and assist with comparisons. By faithfully duplicating all glyphic elements in a modern, readable font, users can compare shapes across hands and also have some assistance in learning how the handwriting of a particular scribe works. The choice of typeface, layout, and other aspects of representation must be informed by aesthetics. Crucially, aesthetics places something as simple as selecting a digital typeface in a historical continuum. Drucker identifies the aesthetic considerations of digital projects as the inversion of the power relationship between formal logic and humanistic aesthetics as a core component of speculative computing. Where old digital humanities projects privileged the role of systems of formal logic over appearance, flipping the roles provides a richer interpretive environment by demanding that formal logic support the needs of aesthetics.[35]
Reusable Components
34> For components to be reusable, they must be documented, accessible, and standard. The techniques of textual editing are fairly well documented but the addition of an apparatus in the form of notes makes sure that the reader understands the process the text went through. Text encoding is, however, only newly documented. Two major technologies are the Medieval Unicode Font Initiative (MUFI) and the Text Encoding Initiative (TEI).[36] Digital tools are preferable within current technology since they are the most reusable, being adaptable to visual print, visual digital, tactile, auditory, and perhaps even gustatory mediums.
35> The Medieval Unicode Font Initiative (MUFI) is a non-profit group of scholars who are supplementing the work of the Unicode Consortium.[37] The Unicode Consortium works on “... developing, maintaining, and promoting software internationalization standards and data, particularly the Unicode Standard, which specifies the representation of text in all modern software products and standards.”[38] (The emphasis is mine.) Unicode is used in almost every recent computer system and specifies a set of encodings with associated glyphs across numerous languages. However, within the conception of the modern lies a narrow definition of characters focusing on recent needs that obscures differences between glyphs and thus collapses anachronistic variant glyphs. For example, a properly encoded Unicode text does not distinguish the sigma form of ‘s,’ from long ‘s,’ from anglicana ‘s,’ or any other variants that do not signify a different underlying letter. These variant ‘s’s are seen by the Consortium as decorative in nature. Initially the Consortium did distinguish historical variants, but decided to stop. In particular, the long-s is a code point in the Unicode standard, but is considered inappropriate for English language long-s. It is present for languages in which the long-s glyph represents a different sort of letter. Therefore, Unicode is intentionally inappropriate for new diplomatic work because it ignores the glyphic components.
36> The MUFI aims to provide tools to distinguish these glyphs as scholars require. They extend the Unicode standard, and provide explication for various encodings, as to what sort of glyph to use for which mark as found in medieval literature. Additional encodings are, of course, essential for diplomatic transcription in which researchers distinguish variants which are only decorative in nature. The work of the MUFI, however, enables one to do distinguish variants in a way that is independent of font choice. Since the standard must specify abstract encodings of glyphs based on Unicode, several different fonts have been developed that support the MUFI standard. In particular, figure 6 uses Andron Scriptor Web which is a free MUFI v. 3.0 compatible set of glyphs to transcribe the beginning of the Monk’s Tale in the Ellesmere Chaucer. (See figure 6.)
Fig. 6, p. 1
Fig. 6, p. 2
Fig. 6, p. 3
Fig. 6, p. 4
Fig. 6, p. 5
Caption: The beginning of the Monk’s Tale diplomatically transcribed from the Ellesmere manuscript.
37> However, a scholar who wanted to reuse data could use another font or even map the font to standard Unicode.[39] The glyphic data is thus encoded, but reusable in a documented way.
38> The Text Encoding Initiative Consortium (TEI-C) develops a standard eXtensible Markup Language (XML) for encoding semantic, physical, and editorial information about all forms of texts. Their scope includes verbal as well as written statements, and supports efforts as diverse as transcription of early Greek papyrus to writing a term paper. The toolbox the consortium created is vast and complicated but extremely well documented which makes it highly reusable for various formats. The TEI-C already provides a set of standard automated stylesheets that can convert texts to HTML, xhtml, DocBook, PDF, RTF, Plain Text, and other formats. Their standard also includes substantial facilities for references and linkages. An image of a page can be linked to the transcribed text and vice versa. The style-sheet can then express the linkage in any way the designer wants: as a hyperlink, an arrow, a written line reference, a quote, embedding the whole text, etc. Thus, the TEI-C provides a abstracted model for encoding texts and critical objects so that the presentation can be adapted to different display systems, platforms, and altered to emphasize different aspects. A TEI-C encoding can be so thick as to be unreadable, while a particular display can show a few aspects that illustrate the argument being made.
39> TEI can also be used to encode the text diplomatically while simultaneously normalizing it and expanding abbreviations. The [g], [glyph], and [char] elements represent various glyphs that can be used when working with a MUFI-compatible font, but ignored for non-MUFI fonts. The [choice] element allows the editor to describe variants in the text as seen in the TEI underlying the transcription of the Monk’s Tale from the Ellesmere Chaucer. (See figure 7.)
Fig. 7, p. 1
Fig. 7, p. 2
Fig. 7, p. 3
Fig. 7, p. 4
Fig. 7, p. 5
Fig. 7, p. 6
Caption: The beginning of the Monk’s Tale diplomatically transcribed in TEI from the Ellesmere manuscript.
40> The encoded text can then be processed automatically to generate the normalized and diplomatic versions. Additions to the work could be to link the text to the digital facsimile and also to link the glossary to the text. Furthermore, the editorial glosses and linkages can be expressed through XML and presented through clever stylesheets. The standard structure of TEI also allows for distributed projects (such as the SDE version of the Hengwrt Chaucer). Each editor, subeditor, transcriber, etc. can be instructed with the same techniques and the final version can be merged automatically into a massive project.
Pedagogical Ladders
41> Providing a glossary and analysis allows the reader to begin to understand what might be going on in the analyzed passage analyzed and both could be linked to the TEI-C’s structure as well as to digital images. The normalized version of the text that can be automatically generated provides a similar facility. A reader who is uncomfortable with Middle English can begin with the Modern English version, read the analysis, and develop some understanding of the normalized version. After working with the normalized version the reader can explore the variant glyphs of the diplomatic version, which adds a new layer of challenge in the form of abbreviations and otiose decorations. Finally, once the reader is comfortable with the diplomatic transcription, she can return to a digital facsimile or the original. Thus, our current study provides a series of hyper- & hypotextually[40]connected steps that enable readers to understand the text better.
Example
42> How can we achieve an effort to exactness in practice using current tools? For the context of this paper, a small passage is marked-up and transformed using various digital tools. As we have been deeply concerned with new directions in diplomatic transcription that digital reproduction facilitate. The coding enabled two versions to be created automatically, the diplomatic transcription and the normalized transcription. These versions can be seen in figure 6 which has both the diplomatic and normalized transcription, while figure 7 has the raw XML describing the content. Both documents repay close reading, in particular the use of [charDecl] which emphasizes a particular effort at exactness in glyphs. The element is used to declare particular glyphs that are not represented in the underlying encoding. Yet, it also offers the opportunity to provide equivalents for display, simultaneously recognizing the impossibility of exact reproduction but the desirability of creating some displayed form. As compared to the digital facsimile, it allows the author to be self-aware about the ambiguities of interpretation, and demonstrates the nature of the choices made.
Conclusion
43> Diplomatic transcription and digital images as both efforts at exactness remain problematic. Their historical approaches and functional goals must remain in the foreground for particular projects as these efforts always result in a new reproduction. Intertextuality is the key to understanding the difference between mindless digital reproductions and the new diplomatic transcription. We have examined several historical attempts at exact copies and discussed some new technological tools for future diplomatic efforts based on how these historical attempts have succeeded and failed. The narrative paints a picture of striving for textual stability against editorial limitations, the limitations of technology, and the limitations of readers. While Chaucer seems to have a firm place in the history of English poetry, the stability of his text still seems questionable ever after hundreds of years of study. New technologies will undoubtedly add to the array of tools that we apply to the very old problem of explaining and establishing a text, and will also lead to new editorial work. To find the way forward we must look backward or otherwise face making the mistakes of our predecessors who attempted exact reproduction.
Notes:
[1] Geoffrey, Chaucer, The Works: Compared with the Former Editions, and Many Valuable Mss. ed. John Urry (London: B. Lintot, 1721).
[2] Gottfriend Wilhelm Leibnitz, Codex Juris Gentium Diplomaticus (Hanover: S. Ammonii, 1693) and also Oxford English Dictionary Online, s.v. “diplomatic,” accessed February 16, 2011,
http://oed.com/view/Entry/53206
[3] The terms aethesis and mathesis, along with the technique are taken, in part, from the work of Johanna Drucker who used similar techniques in developing SpecLab. See: Johanna Drucker, SpecLab: Digital Aesthetics and Projects in Speculative Computing (Chicago: University of Chicago Press, 2009).
[4] This and other primary source materials consulted are in the bibliography at the end; Chaucer, Geoffrey. A Six-Text Print of Chauncer's Canterbury Tales in Parallel Columns from the Following Mss: 1. The Ellesmere 2. The Herzgwrt 154 3. The Cambridge Univ. Libr. Eg. 4. 27. 4. The Corpus Christi Coll., Oxford 5. The Petworth 6. The Lansdowne 851. Edited by F.J. Furnivall. London: Chaucer Society, by N. Trübner, 1868.
[5] The Ellesmere is the well know manuscript, “Ellesmere Chaucer.” EL 26 C 9. Huntington Library, San Marino, Calif., f. 169.
[6] Ideal copy is a fraught term used in bibliography and editorial theory. John Carter provides a good summary of the resonances--both positive and negative--evoked here: “This term, once popular among textual bibliographers, arose from the fact that books printed in the hand-press period (more rarely after) might be corrected during the course of printing, thus creating a moving target, difficult to strike at the ideal moment. Although it is possible in theory for an individual example of the book in question to conform to it, exhibiting the final intention of the author, publisher and printer at the completion of printing, in so far as intention is capable of being established, the ‘ideal copy’ is a sort of Platonic archetype’ laid up where neither moth nor rust can corrupt it. In fact, there ill inevitably be several ‘ideal copies’, distinct in the circumstances of their issue ... But the term has now fallen out of fashion, those concerned preferring to chart the conventional signs of change than pursue a snark that may be a boojum.” John Carter and Nicolas Barker, ABCs for Book Collectors, 8th ed. (London: British Library, 2004): 126-7. Numerous bibliographers and editors have written about the usefulness of the concept of ideal copy and opinions vary wildly.
[7] An older tradition of book-sniffery (here alluded to as bibliophila) can be found in the work of Thomas Frognall Dibdin, Edward Newton, and even occasionally William Blades. A classic of the bibliophilic ilk is Thomas Frognall Dibdin, Bibliomania, or, Book madness : a bibliographical romance, (London: Longman, Hurst, Rees, Orme, and Brown, 1811) A more modern, measured, approach to the aesthetics in a disciplined way is discussed in G. Thomas Tanselle, Bibliographical Analysis: A Historical Introduction (Cambridge: Cambridge University Press, 2009) 81-88, 158, provides a summary of the literature.
[8] This well-trod idea is traced ably in Marshall McLuhan, The Gutenberg Galaxy: the Making of Typographic Man (Toronto: University of Toronto Press, 1962) although the particulars are something of a matter of debate.
[9] Geoffrey Chaucer, The Works of Geoffrey Chaucer, Now Newly Imprinted, ed. by F.S. Ellis (Upper Mall, Hammersmith: Kelmscott Press, 1896).
