Editorial Policies

Transcription Principles and Policies

The core of The Complete Letters of Willa Cather is verified transcriptions of letters encoded in Text Encoding Initiative (TEI) P5 XML markup following the schema developed by the Willa Cather Archive. This encoding standard, which is the best practice in the field, allows the editors both to record detailed information about each letter in the metadata and to construct the edition in the format best suited to long-term preservation.

Three different categories of content have received special attention in Cather's letters: people, places, and works. Each of these categories have authority files, following established TEI protocols, which allow us to connect irregular mentions in Cather's letters to regularized forms. Furthermore, the authority files link content from the edition to international authority services: the Virtual International Authority File (VIAF) and the Getty Thesaurus of Geographic Names. This effort will allow our edition to better integrate with other digital editions that follow the same protocols, increasing the usability in aggregation tools. Additionally, we will one day be able to draw in data from these authority services—like geographic coordinates—and use it in our representation of Cather's letters (for example, creating interactive maps of places she wrote letters from).

The preparation and publication of transcriptions of Willa Cather's correspondence has two distinct steps:

1. Complete markup in TEI XML
2. Transformation of XML data to HTML for display in a web browser

The TEI XML markup is the core data created for this scholarly edition. The reading interface, based on this data, presents a comfortable reading view and a representation of most of the data. Readers can access the XML files from links in the reading view of each letter.

Basic Structure of XML Files

Each individual piece of correspondence is one XML file in the edition, and each file contains a header with information about the correspondence, and body with a transcription of the document.

Each header records the following information about the correspondence:

• Addressee
• Date of composition
• Place of composition
• Repository holding original document
• A brief description of the physical characteristics of the document, including description of accompanying materials
• A list of all hands present in the document, including distinctions between Cather's autograph writing and typewriting
• Information about our digital edition, including personnel and date of file creation
• References to related authority files
• List of digital images related to the item

The header is highly regularized and consistent across the edition. The body, however, varies more broadly, based on the text of the individual item. We have kept extensive guidelines for our markup to train staff members engaging in markup and to keep a record of decisions we make as we encounter new phenomena during the editorial process.

Our guidelines for markup in the body of the letters represent our effort to record as much valuable information as possible in each item of correspondence. Categories of valuable information we record include:

• Formal structure (datelines, salutations, paragraphs, lists, signatures, postscripts, notes, greeting card and postcard text, envelopes, etc.)
• Visible revisions to text (deletions and additions)
• • Note: We do not encode instances in typewritten letters when individual “incorrect” characters are over-typed by “correct” characters—as when Cather noted a typographical error while composing and immediately overstruck a correction with the right letter. These typographical corrections are trivial as Cather was a fairly careless typist and made many errors that she immediately corrected while typing. We do note when entire typewritten words are overstruck with other characters—like a series of “x”s—to indicate deletion, as we've determined these to be more meaningful evidence of her letter composition process.
• Changes in hand (Cather's handwritten additions to a typewritten letter, text by hands other than Cather)
• All references to people, places, and works (linked to authority files), except in the rare cases that we are unable to identify a reference

Because we provide readers with high-quality images of each page of correspondence, revealing the material details of the documents visually, we do not use markup to record details of paper type and dimensions, ink type, and spacing of structural elements on the page.

Text Transcription Guidelines

Our policy is to transcribe all text on each piece of correspondence that is contemporaneous to the original letter composition or receipt. This includes:

• The handwritten or typewritten content of the letter, including revisions (with one minor exception; see above)
• Text printed on the letterhead selected by Cather
• Text printed on the postcard or greeting card selected by Cather
• Text added by another person at roughly the same time the letter was composed or received (such as initials, notes, and date received stamps made in the offices of her publishers)
• Text on envelopes accompanying the original letter, including postmark stamps
• Page numbers and other page order indicators (like “see other side”) written by Cather

We do not transcribe:

• Notes added to the document at a time not contemporaneous to the original letter receipt or composition and therefore not part of the meaning of the original communication (such as notes by archivists or others in possession of the documents)
• Text of accompanying materials written by others and sent with the original letter (such as newspaper articles or letters by others)
• Text written by an unknown hand at an unknown time on a letter or envelope when that text is unrelated to the content of the letter (such as mathematical equations or doodles written on the backs of envelopes); we do transcribe notes that are directly related to the content

Reading Interface Guidelines

Our interface is created through transformation of the core XML data into HTML, which is styled using CSS and JavaScript. The interface is designed to present a clean, comfortable reading interface for a wide array of readers. In the styled reading interface we do not try to provide a facsimile of the original document or to recreate the precise layout and material characteristics of the original text; those qualities are best perceived by the interested user in the accompanying image of the document.

We present a transcription of the letter presented with a regularized formatting that follows Cather's dominant patterns (i.e., dateline at the top right, opening salute at the left, closing salute and signature at the bottom right). If an envelope is part of the original document, it is included after the transcription. Core information about each letter follows the transcription.

Our interface allows users to choose a “plain text” or see a more visually-complicated view that exposes letterhead, notes in others' hands, page breaks, overwritten text, and other details. The default view, clearly indicated in the interface, is the “plain” view. Every reader has access to the more complex view and to the core XML data, but we anticipate that most readers will want to experience the letters through a comfortable, “plain” interface. To be clear: the “plain” interface does not “correct” Cather's text, but merely refers to interface design choices that hide details captured in the markup—like page breaks and repetitive letterhead on successive pages—that interrupt reading flow.

Additions, deletions, uncertain readings, and notes are presented in a way designed to indicate their presence without confusing a reader of the text.

• Deletions made by strikethroughs (by far the most common) are presented as transcribed text with a strikethrough mark: deleted text. Deletions made by overwriting (when words or letters are written on top of other words to delete the first and add the second) are represented with a gray background color
• Additions made above the text line are superscripted; additions made in the margin have partial borders to visually distinguish them; additions made inline are represented with a gray underline
• Autograph additions and revisions made by Cather to an otherwise typewritten letter are rendered as italic
• Areas with illegible text (missing or deleted so heavily as to be unreadable) are noted with “[illegible]”
• A word or passage that cannot be transcribed with absolute certainty because it is at least partially illegible, marked up with the element in XML, is rendered with a bracketed question mark at the end of the passage, such as “Bornea[?]”
• Notes in hands that are not Cather's (such as notes made by recipients) are rendered as gray, italic text
• Printed text (such as letterhead, postcard text, greeting card text, etc.) is rendered in a different font than the letter text. Letterhead and stamped text (typically stamped "received by" dates) appears in gray, while printed text that is considered part of the message (such as greeting card text) remains black.
• In presenting typed letters, we do not visually distinguish between letters typed by Cather personally and those typed for her by a secretary (we do record the identity of the typist, if known, in the markup)

Annotations and regularizations are available as popover notes throughout and at the bottom of the page.

Annotation Policy

Because Cather's letters are so allusive—to people, artistic works, events of the day, and places—informative annotations are particularly important to this edition, and each letter is fully annotated. We provide concise factual, rather than interpretive, annotations, as editorial apparatus is not the appropriate place for critical commentary. We focus our annotation research on subjects not broadly known by most readers, such as Cather's family and personal friends. We only lightly annotate well-known subjects, like Julius Caesar or the Titanic, and those annotations focus on Cather's connection to the subjects. Given the electronic environment of our publication, we intentionally avoid extraneous annotation of material that can be found very easily through a simple Internet search except as it relates specifically to Cather.

Three categories of annotations—people, places, and works—are handled distinctively within the edition. Biographical annotations are written in such a fashion that they can be repurposed across multiple documents. A core annotation is available at every mention of a name in the corpus, and it includes brief biographical details for that person and a succinct description of Cather's relationship with them. In many cases, we also include a photograph of the person in the annotation. Place annotations allow users to see the regularized forms, and works have a simplified bibliographic entry. These foundational annotations are facilitated through our authority file system, described in the Introduction to the edition. For every letter that requires more specific explanation about one of these categories—for example, when knowledge of a specific episode in a person's life illuminates a letter, or when Cather's relationship to a particular work needs to be made clear—we write individualized annotations attached to specific documents.