Contents of this page:
The title "System Requirements" is traditional, but that your system be capable of running a modern graphical (Web) browser is more important than any other detail, so browser requirements would be more accurate. "Modern" means Netscape Navigator 3.0 or higher (including Netscape Communicator) or Microsoft Internet Explorer 3.0 or higher, or any other browser that understands frames, tables, JavaScript, and HTML character entities; can display .GIF and .JPG graphics files; and can play .AIF sound files. If these features are present in your browser and your system runs the browser, you're home free. The CD-ROM is in ISO 9660 format, so it should be possible to use it on any CD-ROM drive in a Macintosh, Unix, or IBM-compatible computer. (Some earlier versions of the Macintosh operating system that support CD-ROM drives will only read ISO 9660 disks with some tinkering--see your system manual for information.) For best results, set your browser to accept the colours for background, text, links, and visited links that are specified in the files, and not to underline links.
Some of the Middle English manuscripts whose versions of the Book of the Duchess are presented here, and the critical edition and reading edition texts that derive from them, use two characters that are not present in the modern alphabet used for English. One of these characters, named "yogh," is not present in any but very specialized fonts, so it has been represented using the number "3," which looks a bit like it (this is a standard ploy from the days of typescript). The other character, named "thorn," is a character still used in the Icelandic language, so it is present in most standard fonts. It is represented in the underlying HTML code on this CD using the entities "þ" (for lower case) and "Þ" (for upper case). The browsers for IBM-compatible and Unix computers interpret these HTML entities correctly and display the letter "thorn," but the Macintosh versions either display instead an inappropriate character (for example, the ligature of f and l) or a little box indicating an undisplayable character. The best discussion of the problem and its solution (there are several options) is on Cathy Ball's Old English Pages at http://www.georgetown.edu/cball/oe/oe-fonts.html. Of the free-download fonts that Dr. Ball discusses, the best for the reading edition text would be Kermit Font A; the best for the critical edition and transcriptions would be Dr. Ball's own Times OE.
It is hoped that navigation of this edition will be relatively easy for people who are
familiar with Web-browsers generally, and that any mysterious features will yield to experiment.
The two main entry points are "titlepge.htm" and "mainmenu.htm".
Use of the
The "Reading Edition" of the Book of the Duchess presents a text of the poem (identical as far as possible with the "Critical Edition" text) on the left side of the screen, together with glossary and explanatory notes on the right hand side of the screen. Most words in the text of the poem are clickable links. Clicking on a word in the poem will bring the glossary definition for that word into the top frame on the right hand side. The explanatory notes are also brought into the bottom right hand frame by clicking, in their case on an asterisk to the left of the line to which the note refers. Sound files and manuscript transcriptions can also be accessed from the reading edition. Each of these elements will now be discussed separately.
The reading edition, as mentioned, is identical with the critical edition, which in turn is based on Bodleian Library MS Tanner 346. The spellings are therefore, in the main, those of that manuscript, and include the Middle English letters "thorn" (þ) and "yogh" (this edition uses number 3), used by the first scribe. No attempt has been made to reconstruct Chaucer's own spelling, nor have emendations, in the main, been made on metrical grounds. The medieval punctuation, consisting of the virgule (represented by a slash "/" and roughly equivalent to the modern comma), the dot (represented by the period ".", and used to mark off numerals and once to separate the words of different speakers), and the paraph mark (represented by the paragraph sign "¶" and used to mark off sections of verse), is conserved. Manuscript capitalization is used. The reading edition has been divided into chunks of 200 lines in length to improve the performance of links to other documents on the screen. This is awkward and will be eliminated when processor speeds allow. A JavaScript utility lets the reader skip to a particular line from the beginning of the poem by entering the line number in a box and clicking a button.
Clicking on most words in the reading text will bring the glossary definition to the right-hand upper frame. (High-frequency words, especially if not carriers of lexical meaning, such as the conjunction "and" or frequently-used prepositions like "of" and "to," are not linked to the glossary after the first few occurrences.) Glossary entries usually have the dictionary form (singular for nouns, infinitive for verbs, uninflected form for adjectives) as headword, in the first spelling used in the base manuscript, though nouns and verbs which only occur once in the poem may have as headword the form that appears in the text. Each glossary entry contains the headword, followed by all of the forms that actually appear in the text with line numbers; contextual definitions, often augmented with related definitions that may have relevance to the text; and the headword of the relevant entry in the Middle English Dictionary (ed. Hans Kurath, Sherman Kuhn, John Reidy, Robert Lewis [Ann Arbor: University of Michigan Press, 1954 - ]. (If plans proceed to place the MED online, the MED headwords in the glossary will become links to the online entry, accessible through the Internet. The new glossary file will be made available for free download at the Web site associated with the CD-ROM.) Within each glossary entry, line numbers are also clickable links. Clicking on one of these will bring the text in the left hand frame to the relevant line; in this way, studies of the poem's vocabulary and Chaucer's use of particular words are facilitated.
The explanatory notes are brought into the right-hand lower frame by
clicking on an asterisk to the left of the line that the note relates to (or first line of a passage that
the note relates to). These notes contain commentary on such things as the allusions and language
used in the poem, the poem's sources in Latin and Old French works, and interpretive and critical
problems. As is always the case with modern editions of Chaucer, the explanatory notes rely
heavily on previous scholarship. An attempt has been made here to give the source of every note
that simply repeats information provided by an earlier edition of the poem--such attribution is not
common practice (perhaps because attributions can clutter the text of notes) but seems like a
reasonable one. When an explanatory note refers to a document that is present on the CD, the
reference is a hypertext link. Clicking on the link will bring the relevant section of the document
to which it refers into the notes frame. (To return to the note from which the document was
accessed, it is necessary use the browser's
The sound files contain a full reading of the poem in Middle English. Recording was done at the University of Calgary during November 1995 with Greg Phillips as recording engineer; the reader was Murray McGillivray. The sound files are in .AIF format and can be played by clicking on the vertical bar ("|") which appears to the left of the reading text adjacent to each line that begins a recorded section. Some browsers require further action (clicking of a tape-recorder-like "play" button symbolized by a right-pointing arrow) to play the sound; on most browsers it is necessary to close the little sound-player window by clicking on its close button to avoid slowing down system performance by keeping too many windows open, since each sound file that is accessed will cause a new window to be opened. In Windows 3.1 especially, since sound-player windows will open directly on top of each other, it is possible to accumulate quite a pile of them without noticing that more than one is open. The reading in the sound files is meant to be illustrative of the general sound of Middle English only, not to provide any kind of scientific accuracy with regard to such matters as dialects, close and open variants of vowel sounds, point or manner of articulation of fricatives, and so on.
The HTML versions of transcriptions of the three manuscripts and the 1532 print can also be accessed from the reading edition. The manuscripts are represented by "sigla," letters (B, F, T, and Þ) which occur in the left margin of the reading text. The appearance of a siglum beside a line indicates the beginning of a manuscript or print page in that witness at that line, except in the case of line 1 and line 97, where the cluster of witnesses beginning pages at the same line has forced sigla to subsequent lines. Clicking on a manuscript or print siglum will bring the transcription of the relevant page of that witness into the notes frame (lower right hand side). Since the transcriptions are also hypertextually linked to the page images, it is possible to bring a page image into the same frame. As when the notes frame is occupied with a source document, it is necessary to click again on an asterisk in the text frame in order to return to the explanatory notes.
The "Critical Edition" has its own introduction, describing the theory and procedures used in its confection and the symbolism used within the text and in the textual notes. The critical text appears in the top frame of the screen and is colour-coded to indicate the degree of manuscript (and print) support for particular readings; the textual notes appear in the bottom frame. The reader must scroll both top and bottom frames; an obvious and planned improvement is to key the textual notes into the critical edition using hypertext links in the style of the reading edition's explanatory notes. When a version has been produced implementing this change, it will be posted on the Web site associated with the CD-ROM.
Full transcription of each of the witnesses is provided in HTML form. The accuracy of these transcriptions is limited both by the competence of the general editor and the presentation medium itself. With respect to the former, the less said the better, perhaps. With respect to the latter, it is worth remarking that the electronic medium and printing methods that derive from the application of electronic technology can be made to implement a far greater variety of graphic representations than would have been practical before our present era, but that by and large this potentiality has yet to be explored in the representation of manuscripts. It is theoretically possible to have a character to represent each of the distinct graphic signs (graphemes) used by a scribe, and thus to present a screen or print image of a manuscript that transcribes that manuscript at the level of the grapheme rather than at the level of the "letter," for example using a different typography for different forms of the letter "s." To do so is desirable, since much detail is lost when distinctions of letter form observed by a scribe are ignored in transcription, when abbreviations are expanded (with or without notice), and so on. The SGML format transcriptions provided at the Web site associated with this CD attempt to provide transcription at the level of the grapheme, though no utility is yet implemented to display these transcriptions in typographical form; the HTML-format transcriptions provided on the CD itself and linked to the reading edition and the page images do not. Rather, they use some standard print-form techniques for slightly increasing the information they provide beyond the usual alphabet of modern English. For example, the common tricks of italicizing the expansion of an abbreviation and of following a flourished letter variant with an apostrophe are used. Scholars who would like a higher level of accuracy are directed to the Web site, although the transcriptions on the CD, except for human errors, present the manuscript readings at a level of scholarly accuracy that the long reign of the print medium has made acceptable.
Electronic images are provided of every page of the Book of the Duchess in each of the witnesses to the text, Bodleian Library Manuscripts Bodley 638, Fairfax 16, and Tanner 346, and William Thynne's edition of 1532. The copyright to the images of the Bodleian Library manuscripts belongs to Oxford University and they may not be reproduced without authorization from the copyright owner. The images of the Thynne print are .GIF images produced by scanning. The manuscripts were photographed by the Bodleian during the summer of 1996, and the resulting 35 mm. slides were digitized in Calgary using the Kodak PhotoCD process. Two sizes of image are provided for each manuscript page, representing the highest and third-highest resolutions of the five resolutions present for each image on a standard PhotoCD. Both the high-resolution image and the smaller image were produced by importing the relevant PhotoCD file into Adobe Photoshop and saving as a JPEG (Joint Photographic Experts Group) format file. JPEG is a "lossy" compression scheme that deals well with graphics with continuous tones and high pixel depth, such as photographs.
Although lossless compression was considered for the CD, the JPEG compression was found to have very little effect on the usefulness of the images for paleographic work, and high levels of compression were achieved. The highest resolution images on standard PhotoCD are 18 MB in uncompressed size (Kodak's own proprietory compression scheme is used on a PhotoCD itself to reduce the file size to about 5 MB). These images when compressed using the JPEG method take up only 500 KB of space on the present CD, with no observable loss of detail (but some very slight visible artifacting of expanses of similar colour). The higher resolution .JPG images are provided for such detailed paleographic work as can be done without handling the actual manuscripts (and with some hope that the provision of such detailed photographs will reduce the need for such handling, which causes deterioration of the manuscripts); the lower resolution images are usually perfectly sufficient for legibility, and since they load more rapidly have usually been linked into the hypertext in preference to the more massive files, though links are provided from the main menu to the larger files as well. The latter are large enough that some low-end machines (for example, IBM-compatibles before Pentium with less than 16 MB of RAM) may have difficulty loading them at all. For many users, these large images may be easier to consult by invoking a photographic manipulation program (like Photoshop) than by using a Web browser.
Chaucer's main known (or suspected) sources for the Book of the Duchess are included on the CD in the original language, Old French or Latin, with translation into modern English of the parts that are relevant to the Book of the Duchess. Full texts are provided of Guillaume de Machaut's Jugement dou Roy de Behaingne, of the Fonteine amoreuse, of relevant Machaut lyrics, and of Froissart's Paradis d'Amour; there are also portions of the Roman de la rose, of Machaut's Remede de Fortune, of the eleventh books of Ovid's Metamorphoses and the Ovide moralisée, and of the descent into the cave of Morpheus in Statius's Thebaid. In all cases, these texts have been prepared by scanning from standard editions believed to be out of copyright and proofreading against the printed text. The punctuation has often been adjusted, and editions subsequent to the scanned edition have often been consulted, altough few readings have been adopted from them. Some scanning errors almost certainly remain after proofreading, especially those that are difficult for the eye to detect, such as the frequent substitution by OCR software of "c" for "e" and of either "1" or "I" for "l," and I would be grateful for any brought to my attention (at mmcgilli@acs.ucalgary.ca). More correct versions of these texts will be mounted to the Web site associated with this CD as such corrections are made.
The translations, with the exception of the translation of the section of Ovid's Metamorphoses, are all the work of Murray McGillivray. They are fairly rough and ready, and are aimed exclusively at bringing the reader to better understanding of the corresponding section of Old French or Latin text. One peculiarity of the translations of verse texts in this edition is that they are often lineated to correspond to the text being translated; although this sometimes has bad effects on how easy the sense of the translation is to grasp, it improves, I hope, the usefulness of the translation as a key to the original language. Both source texts and their translations have line numbers that are clickable links; using these, it is possible to toggle between text and translation; alternatively, the frame documents at the bottom of the Main Menu can be used to bring text and translation into different windows.
The text files contained in the present CD-ROM are in HTML (Hypertext Markup Language) format, for use with Web browsers. HTML is a relative of SGML (Standard Generalized Markup Language), an ISO standard for interchange of textual (and other) data. SGML was adopted as the basis for the Guidelines of the Text Encoding Initiative, released in 1994, a specification for encoding of texts for linguistic and literary study commissioned by three international humanities and linguistics computing organizations. SGML versions of the manuscript and print transcriptions, of the critical edition, and of some of the source texts, are maintained at the Web site associated with this CD for free download or for use at that site. If your computer has an Internet connection, you can link from the Main Menu to that site in order to obtain the SGML versions of the files, together with the DTDs. The SGML versions of the files, particularly the transcriptions, contain more detailed encoding in some instances than the corresponding HTML documents, but you may need to obtain further software tools in order to deal effectively with them.
There is a Web site associated with this CD at
http://www.ucalgary.ca/ucpress/online/pubs/duchess/The Web site contains a truncated free version (missing images, sound, and a number of text files) of this CD. More importantly for purchasers of the CD, it gives access to updates and additions to the hypertext on the disk, which can be used at the site or downloaded free of charge to the reader's computer. Aside from the correction of any errors which are brought to the editor's attention (at mmcgilli@acs.ucalgary.ca), the Web site gives us the opportunity to improve the operation or content of parts of the edition and to allow readers who have purchased the CD to profit from our improvements. In addition, some files, such as the SGML versions of texts and a (planned) bibliography, will be maintained at the site, which will be linked to relevant on-line resources. Among the improvements foreseen in the near future are the improvement of the explanatory notes in the reading edition and the linking of glossary entries to the on-line version of the Middle English dictionary.