Textual

 

Where are we now - how have we got there?

Early computer applications: "classical" critical editing; the computer as a tool

Tools for Textual Criticism and Scholarly editing have been developed in an era when the computer was used as a tool _ for scholarly research, especially for tasks which could not be performed with traditional means.

The first applications therefore have been the collection of variants by automatic collation and stemmatology, i.e. the attempt to automatically establish the relationships and dependencies between the manuscripts: questions which are of paramount importance for the edition of ancient texts.

For an overview of the early phase, see Hockey 1980 and Oakman 1984 and the bibliography Ott 1984.

One of the basic tools is software for automatic collation. Of the collating tools meeting scholarly requirements, two shall be mentioned here: Peter Robinson's Collate, first described in 1989, running on Macintosh Computers, and the program COMPARE (VERGLEICHE) contained in TUSTEP, available in its present form from 1977. Both programs do not only provide lists for manual inspection, but a data base of variant readings for further automatic processing, be it for genealogical studies or for building critical apparatuses or (esp. for editions centered on textual evolution) for other presentations, both in print and electronically.

Solving questions of stemmatology is, at least for rich traditions, impossible without the help of computers. Peter Robinson reports on successful applications of procedures borrowed from evolutionary biology to establish a stemma. In other traditions, similar attempts seem to be less successful, as insinuated by Kurt Aland (1987) in the preface (p. VII) to the publication of the materials resulting from the collation of all the extant manuscripts at 98 variant locations of the Catholic Letters of the New Testament. - A short overview of other attempts to automatize stemmatic research is given by Hockey (2000).

For editing modern texts, where the author's drafts, manuscripts, typescripts, printer's proofs and multiple editions are available, the research interest concentrates on the genesis of the text. The perhaps best known computer-aided edition of this kind is Hans Walter Gabler's edition of James Joyce's Ulysses (1984). Also there, automatic collation helped not only to eliminate transcription errors during data entry, but also produced the entries in the data base, from which, in addition to the synoptic text presentation showing its evolution, the emendation notes at page bottom, and the historical collation list in the appendix, and the reading text were generated. - Further example are the edition of Ingeborg Bachmann's "Todesarten-Projekt", or the edition of the works of Leibniz (for bibliographic details, see below).

"New Philology"; computer as publication medium

In the meantime, both the focus of editing ancient and medieval texts and the role of the computer have fundamentally changed. The "New Philology" does no longer see different versions of a text as witnesses of a lost original which has to be reconstructed from the variants found in the extant copies: not a reconstructed text, but a diplomatic transcription of an extant manuscript or printing, has to be the basis of an edition.

At the same time the computer has taken over a new role, namely the role of a publication medium - and for many an "editor", it looked very "innovative" to not even transcribe the manuscripts (and therewith to "interpret" them), but to put them in facsimile on the web, possibly with links between the single passages: "there is a theoretical purity in unedited, unreconstructed texts that is comforting to editors" (P. Robinson 2000, p.13), and - following the "post-modern thinking against all forms of authority" (P. Robinson) - to appeal to the judgment of the reader to draw his own conclusions.

In the consequence, a disastrous confusion of terms has taken place. It seems necessary nowadays to stress that "editing" and "making available electronically" are by far not synonyms, at least not in our context.

(The same observation can be made in many digitizing projects by which out-of-print works are made available electronically: often only facsimiles are put online or on CD-ROM, without transcription and very limited searching facilities based on some basic structuring data like page numbers or chapter headings.)

Where are we going - promising new developments

Computer technology has made it possible to overcome the limitations of printed editions: variant readings must no loger be listed "atomistically" in a critical apparatus, but can be presented in their full context; adding facsimiles to the transcription of the original sources can help to clarify difficult cases, like paleographically difficult readings, abbreviations, fragmentary sources.

Encoding standards like XML and TEI help in structuring the text, in exchanging, in publication. Linking techniques allow almost instantaneous switching between edition text, apparatus, transcription and facsimile of the sources, commentary, illustrations etc.

On the other hand, the easy availability of these new possibilities is often used as a replacement for necessary scholarly work. An XML-editor like XMETAL is an editor and a valuable tool to control the tagging, but nothing more: editorial work consists of many steps like textual collation, collecting and classifying the variants, analysing, indexing, sorting (e.g. of letters in chronological order), transforming (e.g., normalizing orthography in medieval texts), and formating for which different tools are needed.

Where we are to go?

Many editorial projects and some recent contributions to the discussion (including papers proposed for ALLC/ACH 2002) suggest it to be necessary to further develop the consciousness among editors of the potential of the computer both as a tool and as a medium for scholarly editing.

  • The computer as a tool for scholarly editorial work should be taken more seriously.

Many use it only as an (intelligent) typewriter which makes data entry and data structuring easier. However, the aim of developing more powerful and more specialized tools is not necessarily only to make the work easier; in particular, those tools can not be a replacement for scholarly effort. Their availability is the precondition for better and more reliable results.

For professional work, the application of those tools can not depend on the question if they are expensive or if they can be used only after a steep learning curve. It is one of the characteristics of professional work that professionals have invested more in specialized tools and in the ability to master them than the (scholarly) "consumer" of their products. The consumer who is not a professional in the same field has the right to expect that these products correspond to the state of the art.

  • The computer as a publication medium and a reading device should be taken more seriously.

An electronic edition should not only be a surrogate of a (pretentedly expensive) book; it must offer additional functionality for screening, searching and anlyzing the text, also for solving questions which have not been thought of (or not been handled) by the editing team. This means that the "naked" maked-up text should be included in the electronic distribution.

The questions of preservation and longevity of the results of editorial work should be taken more seriously. The present technical and organizational conditions of electronic distribution (both online and offline) are by far less reliable than those for print media. This should be taken into account at least for editions for whose preparation much time and scholarly effort has been invested or who for other reasons can not be easily be repeated.

For editions published in print, copyright issues should be tackled carefully and in time with the publisher. Will editions which are not (easily) available electronically play a role at all - before the copyright expires? The results of many expensive editorial endeavours are "buried" in libraries, while electronic copies of out-of-copyright (and out-of-date) texts are used even in academic instruction. Who will then learn to estimate the value of an up-to-date scholarly edition? Who will further pay for the preparation of editions which nobody uses?

Printed sources of information (in reverse chronological order):

  • Robinson, Peter M. W.; Gabler, Hans Walter (eds.): Making Texts for the Next Century. Literary and Linguistic Computing, Vol. 15 No. 1 2000 (Special Issue)
  • Hockey, Susan M: Electronic Texts in the Humanities. Oxford: Oxfort University Press 2000 (Chapter 8: Textual Criticism and Electronic Editions, pp. 124-145)
  • Finneran, Richard J. (ed.): The literary text in the digital age Ann Arbor: Univ. of Michigan 1996
  • Catach, Nina (Ed.) Les éditions critiques. Problèmes techniques et éditoriaux. Actes de la Table ronde internationale de 1984 Paris: Les belles lettres 1988
  • Ott, Wilhelm: Bibliografia: Uso del computer nella scienza editoriale. In: La Critica dei Testi Latini Medievali e Umanistici. A cura di A. d'Agostino. Roma: Jouvence 1984, pp. 203-214.
  • Oakman, Robert L.: Computer methods for literary research (2nd. rev. ed.) Athens: Univ. of Georgia Press 1984 (Chapter 6: Textual Editing with a Computer)
  • Ott, Wilhelm; Gabler, Hans Walter; Sappler, Paul EDV-Fibel für Editoren Stuttgart; Tübingen: frommann-holzboog; Niemeyer 1982
  • Hockey, Susan M.: A guide to computer applications in the humanities London: Duckworth 1980 (Chapter 7: Textual Criticism, pp.144-167)
  • Irigoin, Jean; Zarri, Gian Piero (eds.): La pratique des ordinateurs dans la critique des textes. Paris, 29-31 mars 1978. Paris: CNRS 1979 (Colloques internationaux du CNRS, Bd. 579)
  • Dearing, Vinton A. Principles and Practice of Textual Analysis Berkeley : University of California Press 1974
  • Froger, Jacques La critique des textes et son automatisation Paris: Dunod 1968

Software (in reverse chronological order of availability):

  • Classical Text Editor: The word-processor for critical editions, commentaries and electronic publishing. (1997) http://www.oeaw.ac.at/kvk/cte
  • Critical Edition Typesetter. (1996) http://homepage.ruhr-uni-bochum.de/bernt.karasch/cet see also EDMAC: Critical Edition Typesetting (early 1990s) http://www.ucl.ac.uk/~ucgadkw/edmac/index.html
  • Collate http://www.cta.dmu.ac.uk/projects/collate (first described in: Peter Robinson, The Collation and Textual Criticism of Icelandic Manuscripts, in: LLC 4 (1989) 99-105 and 174-181)
  • TUSTEP: http://www.uni-tuebingen.de/zdv/tustep_eng.html - program COMPARE (first described in ALLC Bulletin 5 (1977) 1, 86-87 Main features described in: Wilhelm Ott, The Output of Collation Programs. In: Advances in Computer-Aided Literary and Linguistic Research, ed. D.A. Ager et al. Birmingham: Dept. of Modern Languages, Univ. of Aston 1979, 41-51; for more details see Wilhelm Ott, Mehr als Kollationshilfe: Automatischer Textvergleich als Editionswerkzeug, in: Albert Heinekamp et al. (eds.), Mathesis Rationis. Festschrift für Heinrich Schepers. Münster 1990, 349-372) - typesetting for critical editions: program SATZ (first English description in Wilhelm Ott, The Emancipated Input/Output, in: Alan Jones and R.F. Churchhouse (eds.), The Computer in Literary and Linguistic Studies. Cardiff 1976, 27-37)

Reports on / titles of editions prepared with TUSTEP are found at http://www.uni-tuebingen.de/zdv/tustep/ed.html Among them, the following ones deserve special mention as methodologically / historically important (in chronological order):

  • Kaufringer-Edition, 1972 (first edition typeset with TUSTEP, basis for very elaborate index volume)
  • Klinger, Faust 1978 (from 1974, automatic linking of apparatuses)
  • Krupp, Mischna-Edition 1977 (multiple collations, original plus "reconstructed" Ms.; 7 apparatuses; mixing latin and hebrew fonts, left-to-right and right-to-left printing)
  • Leibniz (ed. Schepers) 1980 ff. (1 volume of ca. 250 pp. every year; advantage of automatic pagination; sophisticated index generation; autographs available; reliably analyzing, handling and documenting the source material)
  • Joyce, Ulysses (1979) 1984: critical and synoptic edition (automatic insertion of critical marks into left hand text from collation results, automatic generation of right hand text from left hand text; historic collation and textual notes in the appendix)
  • Goethe, Works 1985 ff: "edit once, publish many times" (including pocket edition, e-book, CD-ROM)
  • Reeg, Zehn Märtyrer 1985 (hebrew parallel edition of up to 8 versions)
  • van Arkel, Mödruvallabók 1987 (character repertoire; basis for index and concordance volume)
  • Denzinger, Enchiridion 1991 (synoptic printing of up to 4 versions in Greek, Latin, Translations)
  • Bachmann, Todesarten-Projekt 1995 (4 vol.) (sophisticated genetic apparatuses; collation of successive versions)
  • ---------
  • Heinrich Böll (2001ff) (XML-based; linking, indexing; web based distributed editorial system)
  • Averroes Latinus (Schmieja, Univ. Köln, work in progress; automatic collation; non-latin fonts (esp. arabic)