Guidelines for Editors of Scholarly Editions

with Guiding Questions for Vettors, a Glossary, and an Annotated Bibliography

The Modern Language Association's Committee on Scholarly Editions

September 17, 2003

  • 1. Guidelines for Editors of Scholarly Editions
  • 1.1. Principles
  • 1.2. Sources and Orientations
  • 1.2.1. Considerations with respect to source material:
  • 1.2.2. The editor's theory of text:
  • 1.2.3. Medium (or media) in which the edition will be published:
  • 2. Guiding Questions for Vettors of Scholarly Editions
  • 3. Glossary of Terms Used in the Guiding Questions
  • 4. Annotated Bibliography: Key Works in the Theory of Textual Editing
  • Guidelines for Editors of Scholarly Editions

    1.1. Principles

    The scholarly edition's basic task is to present a reliable text: scholarly editions make clear what they promise and keep their promises. Reliability is established by:
    • accuracy
    • adequacy
    • appropriateness
    • consistency
    • explicitness
    —accuracy with respect to texts, adequacy and appropriateness with respect to documenting editorial principles and practice, explicitness and consistency with respect to methods. The means by which these qualities are established will depend, to a considerable extent, on the materials being edited and the methodological orientation of the editor, but certain generalizations can be made:
    • Many, indeed most, scholarly editions achieve reliability by including a general introduction—either historical or interpretive—as well as explanatory annotations to various words, passages, events, and historical figures.
    • Scholarly editions generally include a statement, or series of statements, setting forth the history of the text and its physical forms, explaining how the edition has been constructed or represented, giving the rationale for decisions concerning construction and representation. This statement also typically describes or reports the authoritative or significant texts, and discusses the verbal composition of the text—its punctuation, capitalization, and spelling—as well as, where appropriate, the layout, graphical elements, and physical appearance of the source material. Statements concerning the history and composition of the text often take the form of a single textual essay, but it is also possible to present this information in a more distributed form.
    • A scholarly edition commonly includes appropriate textual apparatus or notes documenting alterations and variant readings of the text, including alterations by the author, intervening editors, or the editor of this edition.
    • And finally, editors of scholarly editions establish and follow a proofreading plan that serves to ensure the accuracy of the materials presented.

    1.2. Sources and Orientations

    1.2.1. Considerations with respect to source material:

    • Is the date of the material known? For example, in William Blake's The Marriage of Heaven and Hell, because the work itself bears no date, the date and its place in the author's oeuvre have to be inferred, and on such inferences other editorial decisions (decisions based, for example, on authorial intentions, which may vary over time) may depend. More generally, the location of a text in time and place may influence the editorial representation of a text.
    • Is there an author? La Chanson de Roland, for example, took a specific written form after a long life as a heroic poem or poems delivered orally from memory. Folktales, which may or may not originate with individual authors, are usually only known to editors in forms that have been shaped by transmission through communities of performers and listeners. W.B. and Georgiana Yeats claimed to have taken dictation from the spiritual world. Sacred texts are often attributed to divine authors or divinely inspired human authors.
    • Is the author known? Authorship has been one of the most powerful and influential categories of textual criticism, where the "authority" of a text has often been determined by its convenient proximity to a known author writing in a specifiable time and space (traditionally, texts that come from an author's hand, such as an autographic manuscript, tend to have more authority in an edition than texts published after the author's death). When a text (for example, Lazarillo de Tormes) has no known author in the modern sense, or when authorship has been collaborative or communal, or when texts have taken shape over an extended period of time, editorial decisions must be based on other grounds.
    • Is there more than one author? For example, Francis Beaumont and John Fletcher collaborated in writing over a dozen dramatic works between 1606 and 1616, such as The Knight of the Burning Pestle; in addition to working together, these two writers also corrected and collaborated on texts with numerous other playwrights, including Rowley, Massinger, Middleton, and Ben Jonson, making it difficult, if not impossible to assign authorship in some of these works to any one specific individual. Harriet Mill's role in the authorship of J. S. Mill's Autobiography might be labeled co-authorship; Theodore Dreiser sometimes revised his novels on the advice of a circle of family, friends, and associates. Max Perkins might be considered the coauthor of the novelists he edited as an employee of Scribner's—most notably Thomas Wolfe, whose published novels bear little resemblance to the manuscripts that Wolfe turned over to Perkins.
    • If there is an author (or authors), how far back in the process of authorship is source material available? For example, there are no surviving manuscripts or working drafts for the majority of Daniel Defoe's more than 250 works, including his novels, such as Moll Flanders or Robinson Crusoe. The editor must rely instead on printed texts produced during Defoe's lifetime as the earliest sources.
    • Does the author play any other roles in producing the object being edited? For example, Vladimir Nabokov translated his own early works from Russian into English, at a later point in his career; William Blake printed and watercolored his illuminated books with the assistance of his wife Catherine; Charles Dickens became his own publisher, first as an editor of Bentley's Miscellany, then as founder and editor of Household Words and All the Year Round.
    • How many other people are involved in producing the object being edited, and what are their roles? For example, John Wilmot, the Earl of Rochester never published any of his works during his lifetime. Some of his poems were printed without his authority in songbooks and miscellanies and they were widely circulated and preserved in manuscript copies. The subsequent posthumous editions gathered together many of these scattered pieces, but a modern editor must untangle the numerous variations found in the verses as they were collected from these various manuscript and unauthorized printed versions. Another example would be the famously vexed case of James Joyce's Ulysses, drafted in longhand, typed by a typist, typeset by printers who spoke no English, and reset as many as five times, after Joyce's editing of page proofs.
    • Is it important, and is it feasible, to reproduce the material sources in facsimile as part of the edition? A facsimile reproduction of an author's manuscript (or diary, or letters, or draft of an unpublished poem or novel) may make it easier to follow the process of composition than any translation of the manuscript into typographic form. For example, recent editors of Emily Dickinson have argued that something important is lost when Dickinson's "jottings" on scraps of paper are translated to the more familiar form of printed poems. In principle, it would seem always to be desirable to reproduce the source material for a scholarly edition in facsimile, but in print editions it is often impractical, and even in electronic editions it may be too expensive, or it may be impossible for lack of permission.

    1.2.2. The editor's theory of text:

    Editorial perspectives range broadly across a spectrum from an interest in authorial intention, to an interest in the process of production, to an interest in reception, and editors may select a given methodology for a variety of reasons. In very general terms, one could see copy-text, recensionist, and best-text editing as being driven by an interest in authorship—but best-text editing might also be driven by an interest in the process of production, along with "optimist", diplomatic, scribal, documentary, and social-text editing. Social-text editing might also be driven by an interest in reception—as "versioning" and variorum editing might be. And of course, an editing practice that is primarily interested in authorship might very well be interested in production and/or reception—any good editor will be aware of the importance of all of these things. However, when an editor has to choose what to attend to, what to represent, and how to represent it, there should be a consistent principle that helps in making those decisions.
    See the CSE's "Annotated Bibliography: Key Works in the Theory of Textual Editing" for further information on editorial methods and perspectives.

    1.2.3. Medium (or media) in which the edition will be published:

    The decision to publish in print, electronically, or both will have an impact on a number of aspects of the edition, on its fortunes, and on the fortunes of its editor. Some questions an editor should consider in choosing the medium of publication:
    • Is the source material itself manuscript, printed, electronic, or a combination of formats?
    • What is the desired or potential audience for the work? Is there more than one audience? Will one medium reach the desired audience more effectively than another?
    • What rights and permissions are required for publication, and do the terms differ by medium?
    • What kind of apparatus can the edition have, and what kind should it have?
    • Are there standard symbols or methods in a given medium for representing the typography, punctuation, or other textual features of the material being edited (Peirce's symbols, Shelley's punctuation, size-of-letter problems, spacing problems)?
    • What is the importance of facsimile material, color reproductions, multiple versions, multiple states, interactive tools in this edition?
    • How important is permanence or fixity? How can those be attained?
    • Alternatively, is there a possible benefit to openness and fluidity (for example, the certainty that new material will come to light)?
    • Is there a publisher willing to publish in the medium you choose?
    • How important is peer review (and if it is important, how will it be provided)?

    Guiding Questions for Vettors of Scholarly Editions

    Last Revised, 15 September, 2003
    Title vetted: ___________________________________________________
    Edited by: ______________________________________________________
    Date vetted: ____________________________________________________
    Vettor: _________________________________________________________
    For each question listed below, the vettor should enter Yes, No, or Not applicable as appropriate. Indication of whether additional comment on this point is made in the attached report is also required.
    YNN/ASee Report
    I. Basic Materials, Procedures, and Conditions
    1.0Has the editor missed any essential primary or secondary materials?
    2.0Has the editor constructed a valid genealogy, or stemma, of all relevant texts?
    2.1Have you tested the validity of this stemma against the collation data and included your findings in the report?
    3.0Have all transcriptions been fully compared by the editor with the original documents, as distinct from photocopy of those documents?
    3.1If any transcriptions have not been fully compared with the originals, is there a statement in the edition alerting the user to that fact?
    3.2Has someone other than the original transcriber carried out a thorough and complete check of each transcription, whether against original or photocopy of the original?
    3.3Have you sampled the transcriptions for accuracy and included the results of that sampling in your report?
    4.0Have all potentially significant texts been collated?
    4.1How many times have the collations been repeated by different people?
    4.2Have you sampled the collations for accuracy and included the results of your sampling in your report?
    II. Textual Essay
    5.0If the edition under review is one in a series, have you examined textual essays and vettors' reports (if any) from earlier volumes?
    5.1Does the textual essay provide a clear, convincing, and thorough statement of the editorial principles and practical methods used to produce this volume?
    5.2Does it adequately survey all pertinent forms of the text, including an account of their provenance?
    5.3Does it give an adequate history of composition and revision?
    5.4Does it give an adequate history of publication?
    5.5Does it give a physical description of the MSS or other pertinent materials (including electronic source materials, if any)?
    5.6Does it give a physical description of the specific copies used for collation?
    6.0Does the textual essay provide a convincing rationale for the choice of copytext, base-text, or the decision not to rely on either?
    6.1Does it adequately acknowledge and describe alternative but rejected choices for the copy-text or base-text?
    6.2If there are forms of the text which precede the copy-text or base-text, can they be recovered from the edited text and its apparatus?
    6.3If not, is it practical, desirable, or necessary to make them recoverable?
    7.0Does the editor give an adequate account of changes to the text made by authors, scribes, compositors, etc.?
    7.1Are such changes to the text reported in detail as part of the textual apparatus?
    7.2If such changes are recorded, but the record will not be published, has the decision not to publish it been justified in the textual essay?
    8.0Is the rationale for emendation of the copy-text or base-text clear and convincing?
    8.1Are all emendations of the copy-text or base-text reported in detail, or described by category when not reported in detail?
    8.2Are the emendations of the copy-text or base-text consistent with the stated rationale for emendation?
    8.3Do the data from collation support the editor's assertion of authority for emendations drawn from the collated texts?
    8.4If the author's customary usage (spelling, punctuation) is used as the basis for certain emendations, has an actual record of that usage been compiled from this text and collateral texts written by the author?
    8.5Have you sampled the edited text and record of emendations for accuracy, and have you included the results in your report?
    8.6Are emendations recorded clearly, avoiding idiosyncratic and/or ill-defined symbols?
    9.0Does the essay somewhere include an adequate rationale for reproducing, or not, the significant visual or graphic aspects of the copy-text or base-text?
    9.1Are all illustrations in the manuscript or the printed copy-text or base-text reproduced in the edited text?
    9.2If not, are they adequately described or represented by examples in the textual essay?
    9.3Are the visual aspects of typography or handwriting either represented in the edited text or adequately described in the textual essay?
    9.4If objects (such as bindings) or graphic elements (such as illustrations) are reproduced in the edition, are the standards for reproduction — sizing, color, and resolution — explicitly set forth in the textual essay?
    III Apparatus and extra-textual materials
    10.0Has a full historical collation been compiled, whether or not that collation is to be published?
    10.1Is the rationale clear and convincing for publishing a selective historical collation (say one that excludes variant accidentals)?
    10.2Does the selective collation omit any category of variants you think should be included, or include any you think should be excluded?
    10.3Is the historical collation to be published accurate and consistent?
    11.0Are the textual notes clear, adequate, and confined to textual matters?
    12.0Have ambiguous hyphenated compounds (e.g. "water-wheel") in the copy-text or base-text been emended to follow the author's known habits or some other declared standard?
    12.1Have ambiguous stanza and/or section breaks in the copy-text or base-text been consistently resolved by emendation?
    12.2Are both kinds of emendation recorded in the textual apparatus to be published?
    12.3For words divided at the end of a line in the edited text, and stanzas or section breaks that fall at the end of a page in the edited text, can the reader tell how these ambiguous forms should be rendered when the text is quoted?
    13.0Does the apparatus omit significant information?
    13.1Can the history of composition and/or revision and/or the history of printing be studied by relying on the textual apparatus?
    13.2Is the purpose of the different parts (or lists) in the apparatus clearly explained or made manifest?
    13.3Is cross-referencing between the parts (or lists) clear?
    13.4Is information anywhere needlessly repeated?
    13.5Is the format of the apparatus adapted to the audience?
    13.6Are the materials well organized?
    14.0Does the historical introduction dovetail smoothly with the textual essay?
    14.1Has the editor quoted accurately from the edited text in the introduction and the textual essay?
    14.2Has the editor verified references and quotations in the introduction and the textual essay?
    14.3Has the editor checked the author's quotations and resolved the textual problems they present?
    14.4Have you spot-checked to test the accuracy of quotation and reference in the introduction, textual essay, and text; and have you included the results of that spot-check in your report?
    15.0Are the explanatory notes appropriate for this kind of edition, for example in purpose, level of detail, and number?
    15.1Is there a sound rationale for the explanatory notes, whether or not the rationale is to be made explicit anywhere in the published work?
    IV: Matters of Production
    16.0Did you see a final or near-final version of the edition or a substantial sample of it?
    16.1If you did not see final or near-final copy, were you satisfied with the state of completion of the materials you did see?
    17.0Has the editor obtained all necessary permissions, for example to republish any materials protected by copyright?
    18.0If there is a publisher involved in producing the edition, has the publisher approved the content and format of the edition?
    18.1Has the publisher approved the amount of time needed for proofreading?
    18.2Has the publisher approved the requirements of the edition's design?
    18.3Has the publisher approved cueing the end-matter (textual apparatus and notes) to the text of the edition by page and line number (if this is a print edition) or by other unambiguous means (if this is an electronic edition)?
    18.4Has the publisher approved the printer or other production facility's copy requirements?
    19.0Has ultimate responsibility for maintaining accuracy throughout the production process been clearly assigned to one person?
    19.1Are the proofreading methods sufficient to ensure a high level of accuracy in the published edition?
    19.2How many proofreadings are done?
    19.3How many stages of proof are there?
    19.4When a new stage of proof is read to verify changes or corrections, is adequate provision made for ensuring that all other parts of the text have not been corrupted?
    19.5Is there a provision in place for collation or comparison of the first correct stage of proof against the production facility's final pre-publication output (for example, blue-lines from a printer, or text as rendered for final delivery in an electronic edition)?
    20.0If the edition—whether print or electronic—is prepared in electronic files, are those files encoded in an open, non-proprietary format (for example, TEI/XML rather than Word or WordPerfect)?
    20.1Will anyone other than the editor create or edit these files?
    20.2Is the editor directly involved in encoding (for example, in doing XML markup, or in coding for typesetting)?
    20.3If automated processes are applied to the text, is the editor checking the result for unintended consequences?
    20.4If an index or search engine is to be used as part of the edition, will it be checked or tested in detail by the editor?
    21.0Can the edited text be easily republished, excerpted, or repurposed?
    21.1If the edition is printed, is it suitable for photographic reproduction? If it is electronic, does it provide PDF or other print-ready output?
    21.2Will all electronic files used in producing the edition be archived?
    21.3Will a correction file be set up and maintained for correcting the text after its initial publication?
    21.4Is the current state of the correction file available to readers of the edition (on the Web, for example, or on request in printed form)?
    V. Electronic Editions (see glossary for acronymn expansion)
    22.1Does the edition include help documentation that explains the features of the user interface and how to use them?
    22.2Does the edition carry a clear statement of the appropriate re-use of its constituent elements, especially those protected by copyright or used by permission?
    23.0Is the text of the edition encoded in an ISO standard grammar such as XML or SGML?
    23.1Is the XML or SGML applied using relevant community guidelines (e.g., The Text Encoding Initiative Guidelines)?
    23.2If the answer to the previous quesiton is "No," then does the essay on technical methods provide a rationale for departing from community practice?
    23.3Is the edition designed to make its underlying markup (rather than markup that results from a rendering process) available to the reader for examination?
    24.0Is character encoding in the edition done according to an ISO standard (e.g. Unicode)?
    24.1Are rendering or transformation instructions (e.g., stylesheets) encoded in an ISO standard grammar such as XSL?
    24.2Does the edition use ISO standard formats (e.g., JPEG, PNG) for the distribution copies of its digital images?
    24.3If there are time-dependent media elements in the edition (for example, audio or video) are these encoded using ISO standard formats (e.g., MPEG/MP3)?
    25.0Are the distribution copies of multimedia elements (image, sound, video) sufficiently high resolution to allow close study?
    25.1Are the distribution copies of multimedia elements stored at reasonable file-size, given the intended method of distribution?
    25.2Are the sources for those distribution copies archived?
    25.3Are those sources captured at sufficiently high resolution to allow for the future derivation of higher-resolution distribution copies?
    26.0Does the edition have, and does it validate against, a DTD or schema?
    26.1Is the DTD or schema used in marking up the edition adequately documented (e.g., with a tag library)?
    26.2If the edition includes one or more databases, is referential integrity enforced within the database(s)?
    26.3Are the database schema(s) documented?
    26.4Are the stylesheets (or other rendering instructions) documented as to their intended effect?
    27.0Is there a definitive and documented method for determining what constitutes the electronic edition?
    27.1Is there a definitive and documented method for determining whether all the constituent elements of the edition actually exist?
    27.2Is technical, descriptive, and administrative metadata provided for all of the components of the edition, using a library-approved schema (such as METS)?
    27.3If any software has been uniquely developed for this edition, is source code for that software available and documented?
    27.4Has a copy of the edition, its images, software, stylesheets, and documentation been deposited with a library or other long-term digital object repository?

    Glossary of Terms Used in the Guiding Questions

    accidentals: A collective term invented by W. W. Greg and now widely used to mean the punctuation, spelling, word-division, paragraphing, and indications of emphasis in a given text—things “affecting mainly its formal presentation,” as he put it (Greg, 21).[1] Greg distinguished between the accidentals of a text and its words, or substantives (q.v.) Accidentals and substantives are conceptually important for Greg's rationale of copy-text, which assumes that authors are more proprietary about their words than about their accidentals, while typesetters and other agents of textual transmission (copyists, typists, proofreaders, copy-editors) are the reverse. For this reason, at least for an edition aimed at preserving the author's accidentals as well as his substantives, the rationale for choosing a copy-text is first and foremost that, of the available texts, it is the most faithful to the author's accidentals and contains the fewest changes to them by other hands. It is therefore often the first or earliest text in a line of descent, but any author who carefully revised his accidentals (say in the second edition) might oblige an editor to choose that text rather than an earlier one.
    authority: Authority is a property attributed to texts, or variants between texts, in order to indicate that they embody an author’s active intention, at a given point in time, to choose a particular arrangement of words and punctuation. Authority therefore always derives from the author, even when author is defined and understood as co-author, collaborator, or a collective (like the Vorticists). Where the author is unknown or uncertain, her authority will need to be argued. It is even possible to invert the usual pattern and assign authority to agents who produce variants commonly regarded as unauthoritative, such as typesetters, proofreaders, or reprint publishers—-though one hesitates to call such agents "author." However defined, the author produces texts or variants which have authority. Some reprints may be said to have "no authority" because the author had no role in producing them. On the other hand, texts that were set from copy revised by the author are said to contain "new authority," meaning that some of their variants arose from her revision. The authority of a holograph manuscript is usually greater than any typesetting of it, but the manuscript's authority at any given point may be superseded if the typesetting incorporates authorial changes—-a case of "divided authority."
    base text: The text chosen by an editor to compare with other texts of the same work in order to record textual variation between them. Its selection can be to some extent arbitrary, or because it is (among the available texts) simply the most complete. Unlike a copy-text (q.v.), it is not assigned any presumptive authority and may not even be used to construct a critical text, serving instead only as an anchor or base to record textual variants.
    collation: Comparison. A collation is either the record of the substantive and accidental differences between two or more texts, or the act of comparing two or more texts for the purpose of documenting their differences.
    copy text: The specific arrangement of words and punctuation which an editor designates as the basis for his edited text, and from which he departs only where he deems emendation is necessary. Under Greg's rationale the copy-text also has a presumptive authority in its accidentals (that is, the editor will default to them wherever variant accidentals are "indifferent"-meaning not persuasively authorial or non-authorial). But copy-text may also designate texts for which no later variants are possible or anticipated. It is now commonplace to designate a manuscript letter that was actually sent as a copy-text for a personal letter. In such cases, emendations of the copy-text would normally not of the author's subsequent revisions, but solely of elements in the original manuscript that the editor could not, or elected not to, represent in the transcription. Contrary to certain common misconceptions, copy-text does not mean the copy an editor or author sends to the printer, and it need not represent the "author's final intention." Indeed it is more likely to be his first draft than his final printed revision of a text. Its selection is based on the editor's judgment that the authority of its accidentals is on the whole superior to other possible texts he could choose for copy-text.
    digital object repository: A means of storing, retrieving, and administering complex collections of digital objects. If the repository is to meet the needs of scholarly editions, it should have a secure institutional basis (like a university research library) and it should have a commitment to long-term preservation, migration, and access. For an example, see
    dtd: Document Type Definition-the set of rules that specifies how the SGML or XML grammar will be applied in a particular document instance.
    emendations: Editorial changes in the copy-text or base text. These changes may be made to correct errors, to resolve ambiguous readings, or to incorporate an author's later revisions as found in printed editions or other sources, such as lists of errata, assuming for the moment that the editorial goal is to recover the author's textual intentions. Different editorial goals might well call for emendations of some other kind, but they would all still be editorial changes to the copy-text or base text and would under normal circumstances be reported as part of the editor's accounting of what she had done with the available evidence.
    end-of-line-hyphens: Hyphens in a word which fall at the end of a line in a manuscript, or a typesetting, may sometimes be ambiguous. They may be either (a) signs of syllabic division used to split a word in two for easier justification of a line of type (or to fit it on the end of one and beginning of the next manuscript line), or (b) signs that a compound word is to be spelled with hyphens ("water-wheel" or "Jack-o-lantern" if broken after a hyphen at the end of a line might be ambiguous, i.e., intended to be spelled with or without the hyphens). For any source text these ambiguous hyphens require judgment as to how the word was intended to be spelled, and such ambiguities would ordinarily be resolved in the way other ambiguous readings in a copy-text are resolved-by editorial choice, recorded as an emendation (change) in the copy-text. In the text as finally edited and printed, if hyphenation of certain words falls at the end of a line and is therefore ambiguous, the editor should likewise resolve this ambiguity for the reader.
    explanatory notes: Notes devoted to explaining what something means or why it is present, rather than textual notes, which are devoted to explaining why the text at a certain point reads in the way it does, and not in some other way.
    historical collation: A record of variants for a given text over some defined number of editions (e.g., from the 1st through the 7th edition) or some period of time (e.g.,from different impressions of the same edition made between 1884 and 1891). The purpose of historical collations is to put before the reader as complete a record as possible of all variants between a group of texts, from which the editor has had to choose. In the past, but only to save space, historical collations have tended to omit variant accidentals and confine themselves to a record of variant substantives.
    iso: International Organization for Standardization, a worldwide federation of national standards bodies from more than 140 countries, one from each country. ISO is a non-governmental organization established in 1947. The mission of ISO is to promote the development of standardization and related activities in the world with a view to facilitating the international exchange of goods and services, and to developing cooperation in the spheres of intellectual, scientific, technological and economic activity. See
    jpg: JPG (or JPEG, for Joint Photographic Experts Group) is an open, non-proprietary ISO standard (official name ITU-T T.81 | ISO/IEC 10918-1) forthe storage of raster images. For more information, see
    machine collation: Collation by means of a Hinman Collator or other mechanical or optical device, allowing very slight differences between states of the same typesetting to be located visually, without the need for a traditional, point by point, comparison of one text against the other. Machine collation is only possible between different states of the same typesetting.
    modernizing: Changing the spelling or punctuation of a text to bring these into conformity with modern standards, as distinct from the standards at the time of first composition or publication.
    mets: METS stands for the Metadata Encoding and Transmission Standard, a standard for encoding descriptive, administrative, and structural metadata regarding objects within a digital library, expressed using the XML schema language of the World Wide Web Consortium. The standard is maintained in the Network Development and MARC Standards Office of the Library of Congress, and is being developed as an initiative of the Digital Library Federation. For more information, see
    mpeg: MPEG stands for Moving Picture Experts Group, and is the nickname given to a family of International Standards used for coding audio-visual information in a digital compressed format. The MPEG family of standards includes MPEG-1, MPEG-2 and MPEG-4, formally known as ISO/IEC-11172, ISO/IEC-13818 and ISO/IEC-14496. Established in 1988, the MPEG working group (formally known as ISO/IEC JTC1/SC29/WG11) is part of JTC1, the Joint ISO/IEC Technical Committee on Information Technology. For more information see
    png: PNG (Portable Network Graphics) is an extensible file format for the lossless, portable, well-compressed storage of raster images. The PNG specification is on a standards track under the purview of ISO/IEC JTC 1 SC 24 and is expected to be released eventually as ISO/IEC International Standard 15948. See
    schema: XML Schemas provide a means for defining the structure, content and semantics of XML documents. For more information, see
    sgml: Standard Generalized Markup Language, a grammar for text encoding,defined in International Organization for Standardization, ISO 8879. For more information, see
    silent emendations: Editorial changes to the copy-text which are not recorded, item by item, as they occur, but are only described somewhere in the textual essay as a general category of change, and are thus made "silently," without explicit notice of each and every change.
    stemma: A schematic diagram representing the genealogical relationship of known texts (includinglost exemplars) of a given work, showing which text or texts any given later text was copied from, usually with the overall purpose of reconstructing an early, lost exemplar by choosing readings from later extant texts, based in part on their relative distance from the lost source. Stemma may also be used simply to show graphically how any given text was copied or reprinted over time, even if the goal is not to recover an early, lost exemplar.
    substantives: W. W. Greg's collective term for the words of a given text -"the significant . . . readings of the text, those namely that affect the author's meaning or the essence of his expression,"as distinct from its accidentals (Greg, 21).* Under Greg's rationale for copy-text, the authority for substantives could be separate and distinct from the authority for the accidentals, thus permitting an editor to adopt changes in wording from later texts, even though she maintained the accidentals of an earlier one virtually unchanged.
    tag library: A document that lists all of the tags, or elements, available in a DTD, with a brief description of the intended use of each, a list of its attributes, and brief statements identifying elements within which this element can occur, and which elements it can contain. See for an example.
    textual notes: Notes devoted specifically to discussing cruxes or points of difficulty in establishing how the text should read at any given point. Compare "explanatory notes."
    user interface: In an electronic edition, the on-screen presentation of content, including navigational methods, menus of options, and any other feature of the edition that invites user interaction or responds to it.
    variants: Textual differences between two or more texts. These would include differences in wording, or in spelling, word-division, paragraphing, emphasis, and other minor but still meaning-bearing elements, such as some kinds of indention and spacing.
    xml: eXtensible Markup Language, a simplified subset of SGML (q.v.), developed by the World Wide Web Consortium. For a gentle introduction to XML, see
    xsl: XSL is a language for expressing style sheets. An XSL stylesheet specifies the presentation of a class of XML documents (for example, TEI documents) by describing how an instance of the class is transformed into an XML document that uses the specified formatting vocabulary (for example, HTML). For more information, see

    Annotated Bibliography: Key Works in the Theory of Textual Editing

    Bédier, Joseph. "La tradition manuscrite du Lai de l'Ombre: réflexions sur l'art d'éditer les anciens textes." Romania 54 (1928): 161-196, 321-356. Bédier advocates best-text conservatism and rejects the subjectivity of Lachmann's method (see Maas) with its emphasis on the lost authorial text, resulting remarkably often in two-branch stemmata. Instead, Bédier focuses on manuscripts and scribes, reducing the role of editorial judgment.
    Blecua, Alberto. "Defending Neolachmannianism: On the Palacio Manuscript of La Celestina." Variants. Eds. Peter Robinson and H.T.M. Van Vliet. Turnhout: Brepols, 2002. 113-33. A clear position statement by the author of the noteworthy Spanish Manual de Critica Textual (1983) in defence of the neolachmannian method. Blecua argues that stemmatic analysis is superior to the methods based on material bibliography and that only the construction of a stemma can detect the presence of contaminated texts.
    Bornstein, George, and Ralph G. Williams, eds. Palimpsest: Editorial Theory in the Humanities. Ann Arbor: University of Michigan Press, 1993. On the assumption that texts are not as stable or fixed as we tend to think they are, these essays examine the palimpsestic quality of texts, emphasizing the contingencies both of their historical circumstances of production, and of their reconstruction in the present. They mark a theoretical period of transition, shifting the focus from product to process in editorial theory and practice.
    Bowers, Fredson. "Some Principles for Scholarly Editions of Nineteenth-Century American Authors." Studies in Bibliography 17 (1964): 223-8. Concise and systematic elaboration of W.W. Greg's theories, arguing that "when an author's manuscript is preserved," this document rather than the first edition has paramount authority and should serve as copy-text. Bowers' principles for the application of analytical bibliography in an eclectic method of editing, have been most influential in Anglo-American scholarly editing.
    Cohen, Philip, ed. Devils and Angels: Textual Editing and Literary Theory. Charlottesville: University Press of Virginia, 1991. The "increasingly theoretical self-consciousness" characterizing textual criticism and scholarly editing marks an impasse, indicative of a paradigm shift. Assumptions that have been self-evident for several decades are rethought in eight stimulating essays and three responses.
    de Biasi, Pierre-Marc. "What is a Literary Draft? Toward a Functional Typology of Genetic Documentation." Yale French Studies 89: Drafts (1996): 26-58. In a continuous effort to present manuscript analysis and critique génétique as a scientific approach to literature, de Biasi designs a typology of genetic documentation, starting from the bon à tirer moment ("all set for printing") as the dividing line between the texte and what precedes it, the so-called avant-texte.
    Eggert, Paul. "Textual Product or Textual Process: Procedures and Assumptions of Critical Editing." Editing in Australia. Ed. Paul Eggert. Canberra: University College ADFA, 1990. 19-40. Starting from a comparison with new techniques of X-raying paintings, Eggert proposes a valuable ideal for a critical edition that allows the reader to study both the writing process and the finished product.
    Ferrer, Daniel. "Production, Invention, and Reproduction: Genetic vs. Textual Criticism." Reimagining Textuality: Textual Studies in the Late Age of Print. Eds. Elizabeth Bergmann Loizeaux and Neil Fraistat. Madison: University of Wisconsin Press, 2002. 48-59. Ferrer defines the difference between genetic and textual criticism on the basis of their respective focus on invention and repetition. He pleads for a hypertextual presentation as the best way to do justice to the diverse aspects of the writing process.
    Finneran, Richard J., ed. The Literary Text in the Digital Age. Editorial Theory and Literary Criticism. Ann Arbor: University of Michigan Press, 1996. The availability of digital technology coincides with a fundamental paradigm shift in textual theory, away from the idea of a "definitive edition." Fifteen contributions reflect on the shift toward an enhanced attention to nonverbal elements and the integrity of discrete versions.
    Gabler, Hans Walter, George Bornstein, and Gillian Borland Pierce, eds. Contemporary German Editorial Theory. Ann Arbor: The University of Michigan Press, 1995. With its representative choice of position statements, this thorough introduction to major trends in German editorial theory in the second half of the twentieth century marks the relatively recent efforts to establish contact between German and Anglo-American editorial traditions.
    Gabler, Hans Walter. "The Synchrony and Diachrony of Texts: Practice and Theory of the Critical Edition of James Joyce's Ulysses." TEXT 1 (1981): 305-26. The work's "total text," comprising all its authorial textual states, is conceived as a diachronous structure that correlates different synchronous structures. A published text is only one such synchronous structure and not necessarily a privileged one.
    Gaskell, Philip. From Writer to Reader: Studies in Editorial Method. Oxford: Clarendon Press, 1978. In 1972, with A New Introduction to Bibliography replacing McKerrow's, Gaskell had already criticized Greg's copy-text theory, arguing that authors often expect their publishers to correct accidentals. From Writer to Reader zooms in on the act of publication and the supposed acceptance of the textual modifications this may involve.
    Greetham, David C. Textual Scholarship: An Introduction. New York and London: Garland, 1992. Impressive survey of various textual approaches: finding, making, describing, evaluating, reading, criticizing, and finally editing the text, i.e. biblio-, paleo- and typography, textual criticism and scholarly editing. The book contains an extensive bibliography, organized per discipline.
    Greetham, David C., ed. Scholarly Editing: A Guide to Research. New York: The Modern Language Association of America, 1995. The most comprehensive survey of current scholarly editing of various kinds of literatures, both historically and geographically, with elucidating contributions by textual scholars from different traditions.
    Greg, W. W. "The Rationale of Copy-Text." Studies in Bibliography 3 (1950-1): 19-37. This pivotal essay has had an unparallelled influence on Anglo-American scholarly editing in the twentieth century. Greg proposes a distinction between substantive readings (which change the meaning of the text) and accidentals (spelling, punctuation, etc.). He pleads for more editorial judgment and eclectic editing, against "the fallacy of the 'best text'" and "the tyranny of the copy-text," contending that the copy-text should be followed only so far as accidentals are concerned, but that it does not govern in the matter of substantive readings.
    Grésillon, Almuth. Eléments de critique génétique: Lire les manuscrits modernes. Paris: Presses universitaires de France, 1994. Introduction to textual genetics or critique génétique, which was developed in the 1970s and became a major field of research in France. In spite of correspondences with textual criticism, it sees itself as a form of literary criticism, giving primacy to interpretation over editing.
    Groden, Michael. "Contemporary Textual and Literary Theory." Representing Modernist Texts: Editing as Interpretation. Ed. George Bornstein. Ann Arbor: University of Michigan Press, 1991: 259-86. Important plea for more contact between textual and literary theorists by the general editor of The James Joyce Archive facsimile edition of Joyce's works.
    Hay, Louis. "Passé et avenir de l'édition génétique: quelques réflexions d'un usager." Cahier de textologie 2 (1988): 5-22; trans. "Genetic Editing, Past and Future: A Few Reflections of a User." TEXT 3 (1987): 117-33. Genetic editing, presenting the reader with a "work in progress," is a new trend, but it revives an old tradition. The founder of the Paris institute for modern texts and manuscripts (ITEM-CNRS) points out that editing has always reflected the main ideological and cultural concerns of its day.
    Maas, Paul. Einleitung in die Altertumswissenschaft vol. 2: "Textkritik." Leipzig & Berlin: Teubner, 1927; Textual Criticism. Trans. Barbara Flower. Oxford: Clarendon, 1958. One of Karl Lachmann's main disciples, Maas systematizes Lachmannian stemmatics, requiring thorough scrutiny of witnesses (recensio) before emending errors and corruptions (emendatio, often involving a third step of divination or divinatio).
    Martens, Gunter and Hans Zeller, eds. Texte und Varianten: Probleme ihrer Edition und Interpretation. Munich: Beck, 1971. Epoch-making collection of German essays with important contributions by, amongst others, Hans Zeller (pairing "record" and "interpretation," allowing readers to verify the editor's decisions), Siegfried Scheibe (on fundamental principles for historical-critical editing), and Gunter Martens (on textual dynamics and editing). The collection's central statement is that the apparatus, not the reading text, constitutes the core of scholarly editions.
    McGann, Jerome J. Critique of Modern Textual Criticism. Chicago: University of Chicago Press, 1983. Repr. Charlottesville: University Press of Virginia, 1992. Textual criticism does not have to be restricted to authorial changes, but may also include the study of posthumous changes by publishers or other agents. McGann sees the text as a social construct and draws attention to the cooperation involved in the production of literary works.
    McGann, Jerome J. The Textual Condition. Princeton: Princeton University Press, 1991. McGann makes several valuable and innovating suggestions, from the idea of a "continuous production text" to a clear distinction between a text's bibliographical and linguistic codes (in the important essay "What Is Critical Editing?").
    McGann, Jerome J. "The Rationale of HyperText." <>; repr. TEXT 9 (1996): 11-32; repr. Electronic Text: Investigations in Method and Theory. Ed. Kathryn Sutherland. Oxford: Clarendon Press, 1997. 19-46; repr. radiant textuality: literature after the world wide web. New York: Palgrave, 2001. 53-74. Conceived in an expressly revisionist relation to Greg's rationale, McGann's ambitious essay presents the book as a machine of knowledge and evaluates the advantages of hyperediting and hypermedia over editions in codex form. As the earliest hypertextual structure the library organization illustrates the theoretical design of a "decentered text".
    McKenzie, D.F. Bibliography and the Sociology of the Text: The Panizzi Lectures 1985. London: The British Library, 1986. McKenzie extends the scope of traditional bibliography to a broader sociology of the text, including videogames, movies, and even landscapes. This perspective has been a major stimulus to the advancement of the sociological orientation in scholarly editing.
    McKerrow, R. B. An Introduction to Bibliography for Literary Students. Oxford: Oxford University Press, 1927. McKerrow's manual of "new bibliography" reflects the early twentieth-century editorial method which made extensive use of analytical bibliography. The author of Prolegomena for the Oxford Shakespeare was rather averse to the idea of emending the copy-text from other sources.
    Nutt-Kofoth, Rüdiger, Bodo Plachta, H.T.M. van Vliet, and Hermann Zwerschina, eds. Text und Edition: Positionen und Perspektiven. Berlin: Erich Schmidt, 2000. As a younger generation's counterpart of Texte und Varianten (see Martens), this state of the art of current scholarly editing in Germany also includes interesting survey articles on Anglo-American scholarly editing (Peter Shillingsburg) and "genetic criticism and philology" (Geert Lernout; trans. TEXT 14 [2002]: 53-75).
    Parker, Hershel. Flawed Texts and Verbal Icons: Literary Authority in American Fiction. Evanston: Nortwestern University Press, 1984. Starting from analyses of revisions by Melville, Twain, Crane, and Mailer, Parker pleads for more attention to textual composition and the development of (sometimes self-contradictory) authorial intentions, which an institutionalized editorial method is often unable to represent.
    Pasquali, Giorgio. Storia della tradizione e critica del testo. Florence: Le Monnier, 1934. Pasquali criticizes some of the basic Lachmannian principles and proposes to take the history of the witnesses and the scribes into account. The current emphasis on textual tradition in Italian philology is to a large extent his legacy.
    Pizer, Donald. "Self-Censorship and Textual Editing." Textual Criticism and Literary Interpretation. Ed. Jerome J. McGann. Chicago: University of Chicago Press, 1985. 144-61. Pizer emphasizes the social aspects of texts, arguing that even if an author has personally changed his texts under external pressure, it may be more important to present the reader with the censored versions because of their social resonance.
    Reiman, Donald H. "'Versioning': The Presentation of Multiple Texts." Romantic Texts and Contexts. Columbia: University of Missouri Press, 1987. 167-80. Reiman suggests "versioning" (or multiversional representation) as an alternative to "editing." The main purpose of this textual approach is to offer readers and critics the opportunity to figure out for themselves how the work evolved.
    Robinson, Peter M. W. "The One Text and the Many Texts." Making Texts for the Next Century. Special Issue of Literary & Linguistic Computing 15.1 (2000): 5-14 In answer to the question "Is There a Text in These Variants?" which Robinson asked in a previous essay, he argues that a scholarly edition is more than merely presenting an archive of variants. The aim of the editor should be to offer a useful tool so as to allow readers to make the connection between variation and meaning. A critically edited text (presented along with "the many texts") is the best means to that end.
    Shillingsburg Peter. Scholarly Editing in the Computer Age: Theory and Practice. 3rd edition. Ann Arbor: University of Michigan Press, 1996. Indispensable introduction to practical procedures and controversial issues in editorial theory, offering clear definitions in matters of textual ontology and a survey of different orientations in scholarly editing.
    Shillingsburg, Peter. Resisting Texts: Authority and Submission in Constructions of Meaning. Ann Arbor: University of Michigan Press, 1997. The editor's main task, Shillingsburg argues, is to relate the work to the documents and to take responsibility for the integrity of the agency of texts, which is a responsibility to both the author and the social contract. Shillingsburg designs a map with four major forms of textual concern, placing the physical documents at the center of textual and literary theory.
    Stillinger, Jack. Multiple Authorship and the Myth of the Author in Criticism and Textual Theory. New York: Oxford University Press, 1991. Stillinger pleads for a broader conception of authorship to include collaboration as an inherent aspect of creation. Case studies include John Stuart Mill and his wife, Keats and his helpers, Wordsworth revising earlier versions of his texts.
    Tanselle, G. Thomas. "The Editorial Problem of Final Authorial Intention." Studies in Bibliography 29 (1976): 167-211. An author's revisions do not automatically reflect his final intentions. In the case of Typee, Herman Melville was responsible for the changes in the second edition, but they represent his "acquiescence" rather than his intention, according to Tanselle, who is well aware that a reader does not have access to an author's mind and advises to always take the context into account.
    Tanselle, G. Thomas. A Rationale of Textual Criticism. Philadelphia: University of Pennsylvania Press, 1989. In his profound analysis of the ontology of texts Tanselle makes a clear distinction between "work" and "text." A work is an entity that exists in no single historical document. Scholarly editing entails, just like any act of reading, the effort to discover the work that "lies behind" the text(s) one is presented with.
    Thorpe, James. Principles of Textual Criticism. San Marino: Huntington Library, 1972. As an early critic of the principles advocated by Greg and Bowers, Thorpe argues that specific compositional peculiarities and contingencies tend to be left out of consideration.
    Timpanaro, Sebastiano. La genesi del metodo del Lachmann. Florence: Le Monnier, 1963. Rev. ed. Padua: Liviana, 1985. The genealogical study of manuscript transmission originated in New Testament criticism toward the end of the eighteenth century. By reexamining Bédier's criticism regarding two-branch stemmata, Timpanaro does not so much aim to correct them but to understand how they came into being.
    Zeller, Hans. "A New Approach to the Critical Constitution of Literary Texts." Studies in Bibliography 28 (1975): 231-63. In his evaluation of Anglo-American copy-text theory from a structuralist point of view Zeller contrasts the practice of editing an "eclectic (contaminated) text" with German editorial methods, showing crucial differences with respect to the notions of "authority," "authorial intention," and "version."


    [1] Greg, W. W. "The Rationale of Copy-Text." Studies in Bibliography 3 (1950-51): 19-36.